Agriculture & Rural Development Working Paper 10 The World Bank Controlling or Living with Floods in Bangladesh Toward an Interdisciplinary and Integrated Approach to Agricultural Drainage Royal Haskoning First printing/web posting: December 2003 ©The International Bank for Reconstruction and Development Agriculture & Rural Development Department 1818 H Street, N.W. Washington, DC 20433 The findings, interpretations, and conclusions are the author's own and should not be attributed to the World Bank, its management, its Board of Executive Directors, or the countries they represent. Some of the numbers quoted are estimates or approximations and may be revised. The boundaries, colors, denomnations, and other information shown on any map in this volume do not imply on the part of the World Bank Group any judgment on the legal status of any territory or the endorsement or acceptance of such boundaries. The Author(s) Royal Haskoning is an independent, global consultancy firm based in the Netherlands. It focuses on the interaction between people and their environment. Postbus 8520 3009 AM Rotterdam Netherlands info@rotterdam.royalhaskoning.com Cover Art Contents Preface...........................................................................................................................................vii Acknowledgments .........................................................................................................................ix Acronyms and Abbreviations ......................................................................................................xi 1. Introduction and Summary ......................................................................................................1 A History of Dealing with Floods and Drainage....................................................................................1 Large-Scale Flood Control and Drainage Projects.................................................................................1 The Shift toward Smaller Scale Development.......................................................................................2 Impacts of FCD Systems.....................................................................................................................2 From Construction to Management......................................................................................................3 Salient Features of Drainage in Bangladesh..........................................................................................3 Flood Control and Drainage Systems--Main Drainage Issues................................................................4 Institutions .........................................................................................................................................4 2. Water Resources and Agriculture ............................................................................................7 The Major River Systems....................................................................................................................7 Climate ............................................................................................................................................10 3. Introduction to Flood Control and Drainage Systems .........................................................17 Floods..............................................................................................................................................17 FCDI Development...........................................................................................................................19 4. Functions and Technological Elements of Flood Control and Drainage Systems .............25 Physical Elements of Flood Control and Drainage Systems .................................................................25 Development of Flood Control and Drainage Systems.........................................................................36 Typology of Drainage Issues and Systems..........................................................................................37 Lessons Learned from FCDI Development.........................................................................................46 Options for FCD Improvement..........................................................................................................47 5. Drainage and Institutions: A Policy Shift..............................................................................43 The National Water Policy................................................................................................................43 The National Water Management Plan...............................................................................................45 Other Relevant Policies.....................................................................................................................45 Public Sector Institutions...................................................................................................................46 iii iv Nongovernmental Organizations........................................................................................................51 Project-User Committees for Water Management...............................................................................52 Informal Local Water Management Arrangements..............................................................................53 Financial and Legal Aspects..............................................................................................................54 6. Conclusions ...............................................................................................................................59 The Need for Agricultural Drainage...................................................................................................59 Typical Characteristics: Multiple Interests, Functions, and Stakeholders ..............................................59 FCD Has Impaired Adequate Drainage ..............................................................................................60 FCD Scheme Performance Has Been Mixed.......................................................................................60 Policy Has Shifted from Flood Control to Living with Floods..............................................................61 Institutions .......................................................................................................................................61 Participation.....................................................................................................................................61 Glossary.........................................................................................................................................63 References.....................................................................................................................................65 List of Tables Table 1 Summary of main characteristics of FCD systems.....................................................................5 Table 2 Features of the hydrological regions.......................................................................................10 Table 3 Features of the hydrological regions.......................................................................................14 Table 4 Overview of classification systems for agricultural and climate seasons ...................................15 Table 5 Total area cultivated with rice in Bangladesh 1975­98 ............................................................19 Table 6 Depth of flooding F0 to F4 land.............................................................................................22 Table 7 Differences between FCD and irrigation and drainage systems in (semi-) arid conditions ..........26 Table 8 Classification of flood control and drainage schemes..............................................................27 Table 9 Overview of scheme types and sizes ......................................................................................27 Table 10 Analysis of BDWB implemented schemes............................................................................28 Table 11 Bangladesh Water Development Board standard embankments designs..................................30 Table 12 Gate operation: benefits and conflicts...................................................................................33 Table 13 Function of khals in different seasons...................................................................................34 Table 14 Main characteristics of the land forms, FCD systems.............................................................38 Table 15 Hydrological regions: main environmental and drainage characteristics..................................39 Table 16a Characteristics of the different FCD systems.......................................................................42 Table 16b Characteristics of the different FCD systems.......................................................................43 Table 17 Effects of global warming on tides.......................................................................................45 Table 18 Expenditures in the water sector...........................................................................................47 Table 19 Institutional models in the Guidelines for Participatory Water Management............................57 Table 20 Institutional models for managing flood control, drainage, and irrigation................................58 Table 21 Summary of main drainage issues in Bangladesh..................................................................60 v List of Figures Figure 1 Bangladesh and the river catchments......................................................................................8 Figure 2 Hydrological regions ...........................................................................................................10 Figure 3 Flood effect on T. Aman, by district 1998 flood ....................................................................18 Figure 4 Thanas with flood control and drainage infrastructure............................................................20 Figure 5 Expansion of irrigation 1982­95...........................................................................................21 Figure 6 Unprotected flood plains ......................................................................................................36 Figure 7 Flood protection ..................................................................................................................36 Figure 8 Reduction of drainage congestion.........................................................................................37 Figure 9 Optimized water control.......................................................................................................37 Figure 10 Main land forms, flood control and drainage systems...........................................................41 List of Boxes Box 1 The Ganges water treaty between Bangladesh and India ..............................................................9 Box 2 Compartmentalization in Polder C ...........................................................................................54 Box 3 Water charges collected by farmers..........................................................................................55 Box 4 Tendering for irrigation in Naogaon.........................................................................................56 List of Pictures Embankment and road ......................................................................................................................29 Cross-dam erected in a khal...............................................................................................................34 Deep flooding in a haor area..............................................................................................................35 Preface The Bangladesh case offers one of the most illustrative examples of agriculture drainage and integrated water management. The country is located in a huge active delta and has one of the highest population densities in the world. The river and flood plain systems, and the manmade improvements-- embankments, roads, and the agricultural and fisheries systems--serve highly diverse stakeholders in fierce competition for scarce land and water resources. In contrast to many of the semi-arid areas, drainage is not solely intended to evacuate excess (irrigation) water but to manage water levels and quality so as to meet the requirements of all stakeholders. This case study is an attempt to develop a typology that would do justice to these specific circumstances, to the land use forms that have been developed, and to subsequent drainage functions that need to be performed. The institutional environment in Bangladesh are reviewed, and arrangements more conducive to sustainable forms of drainage development are discussed. The study is part of a worldwide comparative study commissioned by the World Bank and Wageningen Agricultural University in six countries: Mexico, Egypt, the Netherlands, Pakistan, Indonesia and Bangladesh. The last two countries represent the case of agricultural drainage in humid tropical environments. The Bangladesh Case Study, conducted in summer and fall 2002, is based on desk research outside the country, a three-week field study in Bangladesh and a three-day workshop with the other case study teams in Wageningen. vii Acknowledgments This country study was prepared as a contribution to the project--Agricultural Drainage: Toward an Interdisciplinary and Integrated Approach, sponsored by the World Bank- Netherlands Partnership Program (BNPP)--Environmental Window for Water Resources Management and the Agricultural and Rural Development Department (ARD) of the World Bank. The activity was task managed by Safwat Abdel-Dayem, drainage advisor (ARD) and coordinated by Peter Mollinga, associate professor, Wageningen University. William J. Oliemans undertook the Bangladesh case study for Royal Haskoning, with support by Alamgir Chowdhury, Jan Bron, and Kees Blok. This country study is one of six parallel studies. The other studies are on Egypt, Indonesia, Mexico, the Netherlands, and Pakistan. Together they will provide the basis for formulating the proposed approach. The Task Manager and the Coordinator provided general direction for the study and contributed comments on preliminary texts. The country study benefited from the critical comments provided during the special workshop held in Wageningen, the Netherlands, October 23­25, 2002, and attended by Bank staff, country officials and representatives of International organizations, including the Food and Agriculture Organization of the United Nations, the International Progamme for Technology and Research in Irrigation and Drainage, and the International Commission for Irrigation and Drainage. The Bangladesh Case Study team is grateful for the support received from the Dhaka World Bank Office and. S.A.M. Rafiquzzaman, in particular, who provided guidance, a wealth of information, and assistance with the desk and field study. The team profited from exchanges with staff from the Bangladesh Water Development Board at policy and field levels, stakeholders, universities, nongovernmental organizations, the Local Government Engineering Department, the Water Resources Planning Organization, and various development projects within Bangladesh. Last but not least, the task managers from the World Bank and the Agricultural University of Wageningen as well as other participants at the three-day workshop in Wageningen provided valuable feedback. ix Acronyms and Abbreviations BADC Bangladesh Agricultural Development Corporation BARC Bangladesh Agricultural Research Council BMDA Barind Multipurpose Development AuthorityBWDB BWBD Bangladesh Water Development Board EIA Environmental impact statement EIRR Economic internal rates of return FAO Food and Agricultural Organization (of the United Nations) FAP Flood Action Plan FCD Flood control and drainage FCDI Flood control, drainage, and irrigation GDP Gross domestic product HYV High-yielding varieties IPSWAM Integrated Planning for Sustainable Water Management KSS Krishi Shamabya Samity LGI Local government institution NCA Net cropped area NGO Nongovernmental organization NWMP National Water Management Plan NWRD National Water Resources Database OFWM On-farm water management O&M Operation and maintenance SRP Systems Rehabilitation project Tk Takka UP Union Parishad WARPO Water Resources Planning Organization xi 1. Introduction and Summary Recent policy development in Bangladesh has been marked by a shift in focus from large-scale infrastructure development in the 1960s and 1970s to small-scale development in the 1980s and, starting in the 1990s, to sustainability, maintenance, stakeholder participation and increased attention to the effects of drainage and flood control on nonagricultural land and water use. Typical causes of drainage problems included the obstruction of rain and river water flows due to the construction of flood control embankments--as well as roads and urban areas--without adequate provision for evacuation of drainage water, and local drainage congestion due to insufficient maintenance of drainage networks and outlets within embanked areas. A more recent development has been the appreciation of the value of drainage waterways and water bodies as environmental systems. A History of Dealing with Floods and Drainage Drainage in Bangladesh, one of the wettest countries in the world, has been an age-old aspect of daily life. In the precolonial days, the zamindars [landlords] would construct levees or embankments to protect specific areas and provide rudimentary drainage facilities. Unlike other parts of British India, where irrigation facilities were developed on a large scale, relatively little irrigation infrastructure was constructed in Bangladesh. Given its location at the mouth of two of the world's mightiest rivers, the Ganges and the Brahmaputra, its abundant rainfall and surface water resources, large-scale irrigation was never considered viable. Land and water management developed into an art in the rural areas. Life was adapted to living with floods, and livelihood strategies were developed accordingly. Unique floating rice varieties were bred, growing with the rising flood water at a speed of up to a foot a day, and rural household economies became extremely diversified to cope with some of the world's most difficult living conditions. Flooding and drainage congestion were accepted facts of life, bringing both blessings and occasional disasters. Large-Scale Flood Control and Drainage Projects Explosive population growth, the introduction of more sensitive high-yielding varieties, and a series of disastrous floods in the 1950s and 1960s preceded the first of a number of foreign advisory missions to develop flood control and later flood control and drainage (FCD) projects on a national scale. The first such mission, called the "Krug Mission" after its leader, took place in 1956, marking the start of a large- scale program for embankment construction on the major rivers and flood plains (Krug 1957). Development was to take place through large-scale infrastructure projects that would harness the potential of the fertile soils, spark agricultural growth, and feed and settle the growing population. A massive campaign was undertaken to protect and develop the coastal areas (e.g., through the Coastal Embankment Project in the 1960s), first emphasizing large, fully embanked areas (polders), later tending toward smaller units. 1 2 Controlling or Living with Floods in Bangladesh The Shift toward Smaller Scale Development The 1972 Land and Water Study, carried out by the World Bank, marked a first shift in approach. Instead of large projects, smaller projects were propagated and rapidly implemented in areas with relatively low flooding depths. Projects such as the Early Implementation projects were formulated, and an impressive expansion of protected areas occurred. The Bangladesh Water Development Board (BWDB) was formed as a separate entity and came to play a central part in the FCD development. In an attempt to coordinate and streamline development, several Water Master Plans were developed, in 1986 and later in 1991, and a Master Planning Organization was set up. Despite the need for intersectoral and basin-level planning and implementation, much development took place through sector and donor- driven projects, with each donor pursuing its own policy. The Bangladesh government sought to attain food self-sufficiency, whereby increased rice production, through FCD projects and the cultivation of flood-sensitive high-yielding varieties were the keys to unlock the country's huge agriclutural potential. In the 1970s and 1980s, a quarter of Bangladesh's development budget was earmarked for FCD projects, and agricultural production within embanked areas was raised substantially. Three years after the 1988 flood, self-sufficiency in rice was attained. The next attempt to promote integrated development was the Flood Action Plan (FAP 1989­1995), a hotly debated venture. The plan, a series of studies and pilots aimed at developing an adequate response to the regularly recurring floods--and droughts was a multidonor initiative after the calamitous floods of 1987 and 1988. Much has been written about the plan, with most of the criticism centered on the initial (overly) ambitious plans to construct huge embankments on the major rivers. Major positive outcomes of the initiative included the formulation of guidelines for partic ipation and environmental impact assessment and a better understanding of flood management and the effect of flood control and drainage on fisheries. Impacts of FCD Systems In the late 1980s and 1990s, a more balanced view began to emerge concerning the impacts and benefits of flood control and drainage projects. Within embanked areas, results were generally positive, especially for agricultural production and increased security from flooding. Where drainage provisions were adequate and access to additional water could be secured in the dry season, more intensive cropping was achieved, housing conditions improved, and services were developed. However, negative impacts inside the polders also emerged, for example, as drainage congestion within the embanked areas,1 an increase in flood depths outside embanked areas due to the confinement of the river and flood plain system, a decline in water bodies, interruption of fish migration patterns, and the general lack of adequate maintenance. As a result, repairs were frequently needed. Recent impact evaluations of FCD performance show that small schemes perform better than large schemes and that FCD development in the coastal areas provides higher returns than in inland areas. Since most where FCD development would be economically viable-- those with low to moderate flooding--have already been developed, the future challenges seem to lie in providing for adequate maintenance and developing integrated water management rather than in developing new facilities for agricultural water management 1The embankment kept the floods out but the monsoon rains in. Introduction and Summary 3 From Construction to Management As a result, new FCD development in rural areas to raise agricultural production, is no longer a governmental investment priority, as stated in the 1999 National Water Policy and detailed in the recently completed National Water Management Plan (Ministry of Water Resources 2001c). According the plan, development would focus on rationalizing existing schemes rather than developing new ones. In every case, a fully participatory approach would be applied and implemented according to the Guidelines for Participatory Water Management. In the meantime, new challenges have come to the fore such as the contamination of groundwater with arsenic,2 protection for the rapidly growing urban centers and industrial zones, and the drying up of permanent water bodies. In addition, siltation of major and minor rivers and creeks has contributed to a reduction in the drainage capacity, especially in the coastal areas. These challenges will grow as a result of the expected effects of global climate change. At the same time, the picture is also emerging of a more sustainable form of flood control, drainage, and irrigation development, with decentralized management, multi-stakeholder organizations, and local-level financing. In the National Water Policy, the government makes a commitment to reform and, especially relevant in the case of Bangladesh, the major donors have also committed themselves to these new policies, guidelines, and long-term management plans. Salient Features of Drainage in Bangladesh Drainage in Bangladesh can be characterized by its physical features, uses, and stakeholders. The Physical Features The drainage that has developed in Bangladesh has been shaped by the country's humid climate and physical features. Bangladesh is situated at the mouth of major rivers. The country is a large, flat, and highly active delta under considerable tidal influence, There are three types of drainage: · Gravity drainage under tidal conditions, practiced in the coastal areas where gravity drainage is possible only at low tide · Gravity drainage under nontidal conditions, practiced in the inland areas and taking place at the end of the monsoon season when water levels outside the embankments recede · Pumped drainage, on a limited scale, generally developed with the dual purpose of irrigation and drainage. Infrastructure development (e.g., in the form of FCD for agriculture) and road construction or urban growth have contributed to drainage congestion by impeding natural drainage patterns. Many of the (earlier) flood protection embankments were not equipped with sufficient outlets while the internal drainage network suffered from inadequate maintenance. In the coastal areas, the development of polders has contributed to siltation of the natural drainage channels and subsequently a diminished drainage capacity. Other factors that influence the drainage capacity in the coastal areas are the natural processes of 2"The Largest Mass Poisoning of a People in History,"International Herald Tribune, July 15 2002. 4 Controlling or Living with Floods in Bangladesh accretion and erosion and the fact that Bangladesh receives an estimated 1.5 to 2.0 billion m3 of silt per year from its upstream neighbors. The impact of global warming may prove of paramount importance to the low-lying areas of Bangladesh. Sea levels could rise at a rate of 10 mm a year while temperatures would increase by approximately 1oC in the next 30 years. Monsoon rains would increase by 10 percent in the same period. Consequently, drainage congestion--especially in the coastal areas--would increase. Multiple Uses, Multiple Stakeholders, Multiple Interests A second major characteristic of drainage in Bangladesh is the occurrence of locally diverging uses and interests with respect to drainage infrastructure and water management. Examples include the differing interests of fishermen and agriculturalists or highland and lowland farmers. "Inadequate drainage" for one group is "water storage" for another. A farmer's "reservoir for irrigation" is another group's "fishing ground." The development of FCD infrastructure affects these relations in complex ways. The construction of embankments has interrupted fish migration routes between the flood plains, river systems, and permanent water bodies, but at the same time has provided an environment conducive to the development of pond fisheries. Flood Control and Drainage Systems --Main Drainage Issues In the case study, a typology of drainage systems was elaborated according to three levels of analysis: the infrastructure elements of the FCD scheme, the land forms that have developed as a result of natural, biophysical conditions and manmade interventions, and the hydrological region. The main defining characteristics of FCD systems and corresponding drainage issues are outlined in table 1. In the new National Water Policy and National Water Management Plan, the contours of a more sustainable form of agricultural drainage development have been sketched. They include : · Improving or rehabilitating economically feasible flood control and drainage schemes, to be undertaken only where adequate maintenance provisions can be ensured · Ensuring adequate drainage provisions in all new development · Incorporating participatory planning and implementation approaches · Providing for local-level management, incorporating the interests of all major stakeholders · Developing fish-friendly structures to improve connectivity between the FCD schemes, flood plain, and rivers · Developing flood proofing and preparedness measures and better adapted crop varieties in areas where rehabilitation or flood control is not feasible or desirable, · Improving the drainage capacity in drainage channels in coastal areas (e.g., by dredging tidal basins). Institutions Since water is a crucial resource, some 35 government agencies are involved in the water sector. The Bangladesh Water Development Board, under the Ministry of Water Resources, is by far the most Introduction and Summary 5 significant organization, handling 85 percent of all expenditure in the water sector. Other key players include the Ministry of Agriculture and associated authorities, and the Local Government Engineering Department. At the local level, key actors include local government institutions such as the Union Parishad, project-user committees, and informally organized groups. Table 1 Summary of main characteristics of FCD systems Land form- Hydrologic Type of drainage Main drainage issues FCD system al region infrastructure Deltaic Southwest, River/sea-facing embankments, Managing drainage under tidal conditions coastal Central. and interconnected khals. Major Controlling water levels inside polder for different polder East and minor regulators, flapgates, needs. sluices, and pipes Controlling salinity; balance interests--shrimp and rice Maintaining drainage capacity outside polders; siltation reduces drainage base. Nondeltaic Eastern Sea (river) facing emb ankments Controlling salinity, balance interests agriculture­ coastal Hills with parallel khals salt production­shrimp growing. polder perpendicular to embankment. Managing drainage under tidal influence with Flapgates and regulators cyclones/tidal surge Preventing erosion of embankments (sea/river). Controlling flash floods from hills Beel Northwest, One main river embankment Managing flooding by monsoon rainfall and runoff North with regulators, khal closure, Balancing beel preservation for fisheries with Central and excavation. Connected drainage for agriculture. beels, a few large and many Maintaining fish migration routes: development of small fish-friendly structures Dealing with drainage constraints due to high river water levels and silting up khals Flood plain Across One or two main river Managing congestion due to insufficient drainage country embankments in network of outlets in embankments and roads except rivers. Many khals in area. Maintaining fish migration routes: fish-friendly Eastern Unconnected beels, main water structures Hills and bodies Maintaining dry season water bodies for fish Northeast Limiting flooding from confinement in unprotected areas caused by embankment increases Hoar Northeast Submersible embankments, Managing flash floods in premonsoon haors, beels, and khals Maintaining submersible or dwarf embankments Timing embankment cutting to balance interests of fishermen, boatmen, and agriculture Financing maintenance is possibly the biggest bottleneck to sustainability. Only a third as much as needed is spent on maintenance, which is a constant problem. Users often make substantial in-kind and cash contributions to repair infrastructure, if it is in their direct interest and control. The government's Guidelines for Participatory Water Management foresee extensive involvement of stakeholders in project planning and implementation as well as management transfer to community organizations and local government institutions. Experiments with these forms of local management have 6 Controlling or Living with Floods in Bangladesh begun, initially for irrigation projects but subsequently for flood control and drainage schemes as well. In view of the diverse interests of the many stakeholders and the need for integrated water management, the model of a water board may be appropriate in the Bangladesh circumstances. 2. Water Resources and Agriculture With about 50 percent of its territory prone to regular flooding and a large part of the country receiving more than 2,000 mm of rainfall annually, it is not surprising that drainage and flood control are prominent concerns in rural and urban life in Bangladesh. The magnitude of the drainage and flood problem becomes clear, consideriing the country's key features: · Nearly 93 percent of the country's surfacewater resources originate outside it borders and are therefore difficult to predict and control. · Most of the country is a low-lying, active delta subject to the natural processes of sedimentation, erosion, and flooding. · Rainfall is concentrated into a four-to-five-month monsoon season. · This densely populated country has few employment opportunities outside agriculture, which employs some 80 percent of the total population. The Major River Systems Most of Bangladesh is located within the flood plains of the Ganges, the Brahmaputra and the Meghna.3 Of the total available surface water, nearly 93 percent flows into Bangladesh from outside its borders. The three major river systems drain into the Bay of Bengal through Bangladesh: · The Brahmaputra River enters Bangladesh from the north, flows south for 270 km to join the Ganges River at Aricha, about 70 km west of Dhaka in central Bangladesh. · The Ganges River flows east and southeasterly for 212 km from the Indian border to its confluence with the Brahmaputra, then as the Padma for a further 100 km to its confluence with the Meghna River at Chandpur. · The Meghna River flows southwest, draining eastern Bangladesh including the hills of Assam, Tripura, and Meghalay in India to join the Padma at Chandpur. The Meghna then flows southward for 160 km and discharges into the Bay of Bengal. Bangladesh's position at the outlet of three major river basins is shown in figure 1. 3And their tributaries, such as Teesta, Dharla, Dudhkumar, Surma, and Kushiyara 7 8 Controlling or Living with Floods in Bangladesh Figure 1 Bangladesh and the river catchments Source: Ministry of Water Resources 2001c. The boundaries, colors, denomnations, and other information shown on any map in this volume do not imply on the part of the World Bank Group any judgment on the legal status of any territory or the endorsement or acceptance of such boundaries. The combined discharge of the three main rivers is among the highest in the world.4 Annually, 1,105,612 km³ of water cross the borders of Bangladesh on average, 85 percent of it between June and October. Around 54 percent (598,908 km³) is contributed by the Brahmaputra, 31 percent (343,932 km³) by the Ganges, and nearly 15 percent (162,772 km³) by the tributaries of the Meghna and other minor rivers (FAO Aquastat 2002).5 Flow from the Ganges River is regulated by the Ganges Water Treaty, signed by India and Bangladesh in 1996. The treaty guarantees each of the two countries a minimum flow in the Ganges, according to an agreed formula (box 1). 4Peak discharges are on the order of 100,000 m³ in the Brahmaputra, 75 000 m³ in the Ganges 20 000 m³ in the upper Meghna, and 160,000 m³ in the lower Meghna (FAO Aquastat 2002). 5There are three barrages across the rivers Teesta, Tangon, and Manu, which are used as diversion structures for irrigation purposes only. The Kaptai Dam in the Chittagong Hill tracts is used primarily for hydroelectric power generation Water Resources and Agriculture 9 Box 1 The Ganges water treaty between Bangladesh and India The sharing between India and Bangladesh of the Ganga/Ganges waters at Farakka by 10-day periods from 1 January to 31 May every year will be with reference to the formula as follows: Availability at Farakka Share of India Share of Bangladesh 70,000 cusecs or less 50 percent 50 percent 70,000 cusecs to 75,000 cusecs Balance of flow 35,000 cusecs 75,000 cusecs or more 40,000 cusecs Balance of flow Subject to the condition that India and Bangladesh each shall receive guaranteed 35,000 cusecs of water in alternate three 10-day periods during the period 11 March to 10 May. · Every effort would be made by the upper riparian (India) to protect flows of water at Farakka as in the 40 years (1949­1988) average availability. · In the event the flow at Farakka falls below 50,000 cusecs in any 10-day period, the two governments will enter into immediate consultations to make adjustments on an emergency basis, in accordance with the principles of equity, fair play, and no harm to either party. · The two governments recognize the need to cooperate with each other in finding a solution to the long- term problem of augmenting the flows of the Ganga/Ganges during the dry season. · Guided by the principles of equity, fairness, and no harm to either party, both governments agree to conclude water-sharing treaties/agreements with regard to other common rivers. · The sharing arrangement under this treaty shall be reviewed by the two governments at five year- intervals or earlier, as required by either party and needed adjustments, based on principles of equity, fairness, and no harm to either party made thereto, if necessary. It would be open to either party to seek the first review after two years to assess the impact and working of the sharing arrangement as contained in this treaty. · For the period of this Treaty, in the absence of mutual agreement on adjustments following reviews, India shall release downstream of Farakka Barrage water at a rate not less than 90 percent (ninety percent) of Bangladesh's share according to the formula until such time as mutually agreed flows are decided upon. · The waters released to Bangladesh at Farakka shall not be reduced below Farakka except for reasonable uses of waters, not exceeding 200 cusecs by India between Farakka and the point on the Ganga/Ganges where both its banks are in Bangladesh. · The Treaty shall remain in force for a period of thirty years and it shall be renewable on the basis of mutual consent. Source: Ministry of Water Resources 2001. 10 Controlling or Living with Floods in Bangladesh Hydrological Regions Water resources planning within Bangladesh is done Figure 2 Hydrological regions on the basis of eight defined hydrological regions, as stated in the National Water Policy. The intricate nature of drainage systems within the country requires planning and management of the nation's river systems within the context of hydrological regions. The principal river systems create natural boundaries for these regions. The hilly areas of the East form another hydrological region. The eight regions are indicated in figure 2 and some salient features of the regions are described in table 2. Source: Ministry of Water Resources 2001. Table 2 Features of the hydrological regions Est. Gross Est. pop. Average Area pop. 2025 Rainfall Region (km2) (mln.) (mln.) (mm) Main Towns Southwest (SW) 26,226 15 23.3 1,655 Khulna, Jessore South Central (SC) 15,436 12 14.0 2,307 Barisal, Bhola Northwest (NW) 31,606 32 42.2 1,739 Rajshahi, Pabna, Bogra North Central (NC) 15,949 27 50.4 1,956 Dhaka, Gazipur, Tangail Northeast (NE) 20,061 15 17.6 3,194 Sylhet, Sunamganj Southeast (SE) 10,284 15 18.2 2,271 Chandpur, Comilla, Feni, Noakhali Eastern Hills (EH) 19,956 9 15.3 2,445 Chittagong, Cox's Bazar Rivers and estuaries (RE) 8,607 2,318 Total 148,125 ~125 ~181 2,360 ~ Approximately. Source: Ministry of Water Resources 2001. Climate Bangladesh has a typical monsoon climate, characterized by rain-bearing winds, moderately warm temperatures, and high humidity. It receives heavy rainfall, some 80 percent of it during the monsoon season, from late May to mid-October. Except in some parts in the West, rainfall generally exceeds 1,500 mm annually. Large areas of the South, Southeast, North, and Northeast receive between 2,000 mm and 2,500 mm, and the north and northwestern parts of the Sylhet area receive between 3,800 mm and 5,000 mm. The extreme climate causes monsoon floods in the wet season and water scarcity in the dry season. Water Resources and Agriculture 11 Climate and the location of most of Bangladesh in an active delta pose major challenges for agricultural water management. Potential evapotranspiration is uniform around the country at 1,300 mm per year (Ministry of Water Resources 2001). Illustrative of Bangladesh's challenging climatic environment is the fact that droughts and floods occur with almost equal regularity; droughts on average every 2.3 years and floods every 1.8 years (FAO Aquastat 2002). Although floods wreak havoc, they are seldom the sole cause of major disasters. The population has managed to adapt to these regular, seasonal floods, which, it is sometimes said, are both a blessing-- through soil fertilization and the provision of breeding grounds for fish--and a curse--through displacement, destruction of houses, and sometimes loss of life. The most extensive flooding generally occurs in monsoon season when rainfall and river flooding coincide, and the river levels do not allow for the drainage of rainfall floods from the adjacent areas. The most disastrous floods, in terms of lives and livelihoods lost, occur in the coastal areas when high tides coincide with the major cyclones (Brammer 1996, Ministry of Water Resources 2001). The Role of Agriculture in the Economy The economy of Bangladesh is heavily dependent on agriculture. About 84 percent of the total population live in rural areas and are directly or indirectly engaged in a wide range of agricultural activities. Agriculture accounts for 31.6 percent of the country's GDP. Of the agricultural GDP, the crop subsector contributes 71 percent, forest 10 percent, fisheries 10 percent, and livestock 9 percent. Agriculture generates 63.2 percent of total national employment; the crop sectors' share of national employment is nearly 55 percent (Ministry of Agriculture, Bangladesh 2002; FAO 2002). Within the crop sector, rice dominates, with an average 71 percent share of the gross output value of all crops. As a result, growth in the agricultural sector largely mirrors the performance of rice production. Bangladesh produces about one fifth of the world's raw jute. Jute and tea, principal sources of foreign exchange, are the most important agricultural products after rice. Other important agricultural products are wheat, pulses (leguminous plants such as peas, beans, and lentils), sweet potatoes, oilseeds, sugarcane, tobacco, and fruits such as bananas, mangoes, and pineapples. Export earnings from fish and fish products, in particular shrimp, are also sizeable (Ministry of Agriculture 2002, FAO 2002). Agricultural Seasons Cropping patterns are governed by the seasonal rainfall variations. From June to September, the Southwest monsoon season brings heavy rain, allowing rice cultivation. From October to May, the inadequate amount and reliability of rainfall hampers crop production, especially in December, January, and February. With irrigation, conditions in the dry season are ideal for growing rice and many other crops. The main cropping seasons are: · Kharif I (March to July), essentially the premonsoon season. Aus is the type of rice planted in this season. Jute is the other principal crop. · Kharif II (July to December), the monsoon and immediately post-monsoon period. Aman rice is the main crop. · Rabi (November to April), the winter season. A much wider range of crops is grown, including wheat, pulses, oilseeds, vegetables, and spices. Boro paddy, transplanted between December and February and harvested largely in May, is the most widely grown crop. The rabi season is the main 12 Controlling or Living with Floods in Bangladesh irrigation season with limited supplementary irrigation of aus and aman also taking place in Kharif (Ministry of Water Resources 2001). Classification systems for the seasons have been developed in Bangladesh, depending on the purpose of the classification. An overview is presented in table 3. The main characteristics of the seasons are (Brammer 1996): · Premonsoon season. March­May, with the highest temperatures and evaporation rates, occasional heavy rainstorms, tropical cyclones in coastal areas · Monsoon season, June-September, the period of highest rainfall (more than 80 percent of annual rainfall) · Post-monsoon season, October­November, hot with occasional thunderstorms, tropical cyclones in coastal areas · Dry season, December--February, relatively cool and dry with a few winter rains. Landownership and Poverty Landownership is heavily skewed toward large landholders in Bangladesh. Roughly 10 percent of farm households own 51 percent of the agricultural land; the bottom 40 percent own only 2 percent (FAO 2002). Poverty alleviation remains one of the main challenges for the country, and successive policies and development programs have set poverty alleviation as the ultimate objective for several decades. Significant economic and social progress has, however, been made over the past 20 years. Real wages increased by 7 percent in 1995­2000 (World Bank 2002), and GDP increased from US$279 in 1990 to US$370 in 1999. Progress has reached the villages, where facilities have also shown a notable improvement the last 20 to 30 years, not least due to a marked increase in agricultural production. Yields have increased at an average of 2.2 percent a year, while cropping intensities increased from 155 percent during the late 1970s to 175 percent in the early 1990s. The main contributors to cropping intensification have been: increased labor input; mechanization; and, to a smaller extent, irrigation (Ministry of Water Resources 2001). The total area under rice can be seen from table 4. To keep up with population growth, the expansion of the past three decades will need to continue. Even with the currently decreasing population growth, down to 1.7 percent in 1998 from 2.5 percent in 1971, and declining further to 1.15 percent in 2025 and 0.63 percent in 2050, Bangladesh will still have a population of 181 million in 2025 and 227 million in 2050, making Bangladesh one of the world's most densely populated countries.6 The projected growth of the population will occur wholly in rural areas (Ministry of Water Resources 2001). 6With an area of 147,570 km2, the present population density is 875 inhabitants per km2; this would increase to 1,230 and 1,540 inhabitants per km2, respectively. Water Resources and Agriculture 13 14 Controlling or Living with Floods in Bangladesh Table 3 Features of the hydrological regions Bengali Bengali season Cropping season Month Climate seasona Climate seasonb Climate seasonc month Jhoistho Grissokal June Summer Ashar Kharif Borsakal (rainy I season) July Shraban Monsoon Aug Monsoon Bhaddro Monsoon Sorotkal (fall) Kharif Sep Ashwin II Oct Kartik Hemontokal Post-monsoon (late fall) Rabi dry Nov Winter Early dry 13 Aghrahayn Rabi Boro Dec Dry season Pous Sitkal (winter) Jan Magh Feb Critical dry Phalgun Bosontokal (spring) Mar Summer Chaittra Kharif I Premonsoon Apr Boisakh Grissokal (summer) May Late dry Jhoistho Sources: a. BBS Bangladesh Bureau of Statistics 2002; b. FAO--Brammer 1996; c. Ministry of Water Resources 2001. Water Resources and Agriculture 15 The economic development achieved in recent years has allowed Bangladesh to lessen its dependence on foreign aid, which now finances less than one sixth of all national investment and less than half of the government's Annual Development Program. The challenge is enormous; in 1998 an estimated 67 million persons, 53 percent of the total population, were classified as poor, with 36 percent living in absolute poverty7. Poverty in Bangladesh is closely associated with landownership. Functionally landless (less than 0.2 ha) households comprise 37 percent of the rural population (Ministry of Water Resources 2001) and 65 percent of the poor (FAO 2002); marginal landowners (with between 0.2 ha and 0.6 ha) account for another 21 percent (FAO 2002). The prevalence of extreme poverty is higher among female-headed households, whose total population may exceed 4 million. More than 95 percent of these female-headed households fall below the poverty line, and their incomes are, on average, 40 percent less than those of male-headed households (FAO 2002). With the expected population increase, this situation is expected to remain highly critical. Table 4 Overview of classification systems for agricultural and climate seasons Crop area Crop area Change Rice crop type 1975­76 1997­98 (x 1,000 ha) (x 1,000 ha) (x 1,000 ha) Aus: High-yielding varieties 353 490 +137 Local 3,068 1,075 ­1,993 Total aus 3,421 1,565 ­1,856 Aman: T. aman, high-yielding varieties 557 2,547 +1,990 T. aman local 3,373 2,448 ­925 B. aman 1,831 841 ­1,017 Total aman 5,761 5,809 +48 Boro: High-yielding varieties 642 2,773 +2,131 Local 506 238 ­268 Total boro 1,148 3,011 +1,863 Total rice 10,330 10,385 +55 1,552 5,810 Total high-yielding varieties (15 %) (56 %) +4,258 8,778 4,575 Total local varieties ­4,203 (85 %) (44 %) Source: Ministry of Water Resources 2001. 7Poverty is defined here in terms of caloric intake. Less than 2,122 calories per day is classified as poor and less than 1,805 calories per day as absolute poor (Ministry of Water Resources 2001). There are other, possibly better, means of defining poverty, for example, related to resource access rather than only caloric intake. 16 Controlling or Living with Floods in Bangladesh Structural Reforms in the Agriculture Sector Not surprisingly, policy in the agriculture and water sectors has been, and will continue to be, driven by the concern for food self-sufficiency, especially in grains (rice). While the drivers of this policy were public sector institutions in the 1970s and 1980s, through FCD development, promotion of high-yielding varieties (HYVs) and subsidized minor irrigation development, the late 1980s witnessed a shift toward the private sector. A program of structural reform was introduced, and state controls and subsidies were gradually phased out. Insofar as drainage and irrigation is concerned, the most important changes were the: · Removal of import restrictions on small diesel engines in 1986­87 and lifting of duties on these imports in 1988­89 · Removal of subsidies on deep tubewells in 1992 · Complete phasing out of the role of the Bangladesh Agricultural Development Corporation (BADC) from procurement and distribution of minor irrigation equipment in 1993 (Ministry of Agriculture 2002). These reform measures had a tangible effect on increasing demand for irrigation equipment and consequently on the rate of increase in the area under minor irrigation. A vibrant irrigation equipment sector developed, greatly enhancing the accessibility of irrigation equipment to the agriculture sector. By 1996, more than an estimated 500,000 wells had been installed. The minor irrigation market is estimated to have reached an annual turnover in excess of Tk 1.5 billion (US$30 million) in sales and repairs alone (Ministry of Water Resources 2001). 3. Introduction to Flood Control and Drainage Systems Bangladesh is the most flood prone country in the world. About 1.32 million ha and 5.05 million ha of the total cropped area is severely and moderately flood prone. Floods Normal flooding (barsha) affects about 30 percent of Bangladesh each year (FAO 2002, Ministry of Water Resources 2001), but land use and settlement are well adapted to it. Abnormal flooding (bonya) can submerge more than 50 percent of the land area, damaging crops and property, disrupting economic activities, and causing loss of life. Four types of floods can be distinguished in Bangladesh · Rainfall floods, occurring throughout the country in the monsoon season and often aggravated by inadequate drainage provisions · River floods, occurring on the active flood plains of the major and minor rivers · Floods due to storm surges, occurring in the coastal areas, caused by tropical cyclones in the pre- and post-monsoon seasons · Flash floods, occurring mostly in the Northeast and Eastern Hills regions at the onset of the monsoon season. The effects of flooding on the main monsoon crop, transplanted (T.) aman rice, can be severe, with up to 45 percent production loss in severely flooded areas (figure 3). Flooding Depth Classification The Food and Agricultural Organization (FAO 2002) classifies the land flooding type according to the maximum depth of flooding, with the "F1" category (maximum depth of flooding 0.9 m) accounting for close to half of all the land prone to flooding. Table 5 shows the distribution of land by flood stage. By excluding flood waters from an area, the basic function of FCD infrastructure for agriculture is to convert F1, F2, or F3 land to a shallower flood phase, which, if successful, facilitates conversion from broadcast aman to T. aman and from local to high-yielding varieties. As a result, yields increase. Another benefit in the inland areas is protection of the boro rice crop from early flash floods, mostly in the Northeastern region. For the coastal areas, the main benefit lies in the exclusion of saline water from the area. 17 18 Controlling or Living with Floods in Bangladesh Figure 3 Flood effect on T. Aman, by district 1998 flood Source: Environmental and GIS Support Project, 1998. Bangladesh Agricultural Research Council. EGIS 1999. The boundaries, colors, denomnations, and other information shown on any map in this volume do not imply on the part of the World Bank Group any judgment on the legal status of any territory or the endorsement or acceptance of such boundaries. Introduction to Flood Control and Drainage Systems 19 Table 5 Total area cultivated with rice in Bangladesh 1975­98 Land type Maximum depth Seasonally flooded Permanently Approximate area of flooding (m) average (%) flooded(%) (km2) Medium Highland 1 (F0) 0.3 16 0 12,124 Medium Highland 2 (F1) 0.9 44 1 34,098 Medium Lowland (F2) 1.8 23 1 18,186 Lowland (F3) 3.0 11 3 10,608 Very Lowland (F4) >3.0 1 1 1,515 Total 94 6 75,772 Sources: FAO 1996; Ministry of Water Resources 2001. FCDI Development Flood control and drainage development has occupied a prominent position in Bangladesh's development effort since the late 1950s. It has long been considered instrumental in achieving the government's objective of increasing food production and increasing security against life-threatening floods. As a result, approximately 6 million ha have been equipped with some form of FCD infrastructure in Bangladesh up to the present day,8 nearly total coverage of the areas susceptible to regular flooding.9 Initially emphasis was placed on flood control and drainage, but later the focus shifted to irrigation. Recent developments emphasize the (re)development of the fisheries function in areas equipped with FCD infrastructure. The Role of Irrigation The expansion of the irrigated area has been spectacular in Bangladesh (figure 5). Minor irrigation10 covered 1.5 million ha in 1982, 2.6 million ha in 1990, and nearly 4 million ha in 1998. The 1997­98 National Minor Irrigation Census reported 3.82 million ha of mostly minor irrigation, most of it private and locally managed (2.80 million ha). Major irrigation, managed exclusively by the BWDB, covered 0.30 million ha, the remainder being traditional unmechanized irrigation and deep tubewells (mostly under public management). Observations after the 1998 flood, when a bumper crop was harvested, may indicate a further increase in irrigated area to 4.3 million ha (Ministry of Water Resources 2001). 8To be precise, 6,104,764 ha have been equipped, according to the latest BDWB inventory (WSIP 2002). 9A substantial part of the net cultivable area (9.15 million ha) has also been equipped. The cultivated area equals some 7.74 million ha (FAO 1999 in Knegt 2000). 10Through low lift pumps, shallow and deep tubewells, and unmechanized means. 20 Controlling or Living with Floods in Bangladesh Figure 4 Thanas with flood control and drainage infrastructure Source:National Water Resources Database. Ministry of Water Resources 2001. The boundaries, colors, denomnations, and other information shown on any map in this volume do not imply on the part of the World Bank Group any judgment on the legal status of any territory or the endorsement or acceptance of such boundaries. Introduction to Flood Control and Drainage Systems 21 Figure 5 Expansion of irrigation 1982­95 4.00 3.50 3.00 2.50 ha 2.00 Million1.50 1.00 0.50 0.00 1980 1982 1984 1986 1988 1990 1992 1994 1996 Source: Ministry of Water Resources 2001. Part of the expansion is made possible because of the existence of FCD infrastructure, which provides increased security and retention and inlet of irrigation water (flushing). Through the construction of katcha11 cross-bunds and the excavation of earth canals, water levels can be better controlled, and infrastructure originally designed for drainage can thus also be used for irrigation. Irrigation inside FCD areas is typically drawn from surfacewater in the khals[lakes in the floodplains] or beels [natural channels] by means of low-lift pumps and hand-powered lifting beams and swing buckets. Typically, the lifting device belongs to an individual farmer, while the conveyance canals are often constructed by groups of farmers. History of FCD Although much of the development is recent, floods and "living with floods" has been part of life in Bangladesh for ages. Before the advent of "modern" government administration, local management of FCD systems was a traditional feature. Well known examples include the construction of submersible embankments by landlords in the Northeast region and the low flood protection embankments in the coastal areas in the Southwest, dating to the 1800s or before. Local flood protection works on the Gumti date even farther back (Ministry of Water Resources 2001). An overview of the main historical developments in FCD is presented in table 6. 11Katcha means "unripe" and pucca means "ripe." These terms are commonly used to refer to "unlined" or "lined" in irrigation, "unpaved" or "paved" in roads, "temporary, farmer-constructed" vs. "permanent, agency-constructed" bunds, embankments, sluices, gates, and so on in FCD systems 22 Controlling or Living with Floods in Bangladesh Table 6 Depth of flooding F0 to F4 land Period Duration Key events Ancient period Up to 1200 A.D. "overflow irrigation" Medieval period Sultani period Construction of embankments, roads, tanks Turkish period 1200­1757 Construction of embankments, roads, tanks Construction of planned canals and embankments, river dredging, Mughal period development of road link and river network Colonial period 1757­1947 Embankment, reservoirs, and tanks gradually decline for lack of maintenance Krug Mission 1956­57, and First Master Plan 1964 Start Coastal Embankment project 1968 World Bank Land and Water Resources Sector Study 1972 Creation of the Bangladesh Water Development Board (BWDB ) and start of small-scale FCD schemes 1972 Modern period 1947­present Establishment of Master Planning Organization 1980 and Water Resource Planning Organization 1991 National Water Plan I 1986, and II 1991 Flood Action Plan 1989 Merger WARPO and FPCO 1999 National Water Policy 1999, and National Water Management Plan, 2001 Source: Adapted from Md. Liakat Ali (2002); additions by authors. The interest in FCD and irrigation diminished in the colonial period when emphasis was placed on navigation for trade. The Mughals had set up an independent department, whereby local landlords, in cooperation with the traditional local councils (panchayats) were empowered to levy land taxes for activities such as river dredging and maintenance and new construction of embankments, roads, and bridges. During the colonial period, this system was formally abolished, although informally, landlords remained occupied with these activities, being the largest landholders in the area. Nevertheless, the lack of central support to panchayats and these types of works started a process of slow decline, which was reversed only in the "modern period" (Md. Liakat Ali 2002). The National Water Plans In modern history, water planning has featured prominently and has spurred much of the large-scale infrastructure development. It started in the 1950s after the disastrous floods of 1954 and 1955 in what was at that time East Pakistan. In 1956­57, a United Nations Mission, led by J.A.. Krug, undertook a major review and recommended the construction of flood control systems on a large scale. In 1959, the East Pakistan Water and Power Development Authority was created and assigned responsibility for water resource (and power) development (Krug 1957; Wester and Bron 1998, Ministry of Water Resources 2001, Md. Liakat Ali 2002). The First Master Plan (1964). Along the same lines, major infrastructure development was proposed in the first Master Plan, drafted in 1964. In the plan, 58 large-scale systems were proposed, covering 5.8 million ha, comprising embankments, pumping stations, and canal irrigation. From the 1960s to the 1980s many of these systems were implemented, though in adapted form and without the pumped drainage originally foreseen. One of the main principles underlying the master plan was that full flood control was the key to increasing agricultural production by excluding river flood water from agricultural land and draining Introduction to Flood Control and Drainage Systems 23 excess rainwater from the embanked areas by sluicing or pumping (EPWPDA 1964 in Wester and Bron 1988). The 1964 master plan was a typical product of development thinking at that time with its emphasis on large-scale infrastructure development and a central role for the public sector. It quickly became clear, however, that the plan severely underestimated local complexities, both social and hydrological. Consequently, the World Bank, in 1966, recommended against its implementation. A major positive impact was the protection of most of the nation's coastal zone against tidal flooding (Ministry of Water Resources 2001). The Land and Water Study (1972). Following the World Bank Land and Water Study of 1972, the emphasis on large-scale works for full flood control diminished, and donors shifted their attention toward getting quick results from development efforts.12 The development of small-scale irrigation, supported by low-cost, medium-sized FCD systems in shallow, flooded areas, was advocated. Nevertheless, the basic premise that full flood control and improved drainage were prerequisites for agricultural development was not questioned, and the study retained the original focus of expanding the area under FCD and attaining food self-sufficiency. The development of medium and later small FCD steadily increased, reaching 2.7 million ha in 1985 and 3.5 million ha in the early 1990s (Wester and Bron 1998), to 6.1 million ha today (Ministry of Water Resources 2001). The National Water Master Plan (1986­91). In the meantime, in 1980, the Master Planning Organization was set up, and the first national Water Master Plan was produced in 1986 and updated in 1991. Both plans were basically a continuation of the 1972 study, advocated the continuation of FCD development and, like their predecessors, were essentially food-grain self-sufficiency plans. Nonagricultural water- using sectors did receive mention but were not included explicitly. Neither the 1986 master plan nor the 1991 update was adopted by the government of Bangladesh. The first major shift was to come about during the period of the Flood Action Plan (FAP). The Flood Action Plan After the severe floods of 1987 and 1988,13 national and international attention again focused on the need for flood control. The government of Bangladesh asked the World Bank in 1989 to coordinate the various donor efforts so as to arrive at a comprehensive flood plan. This request resulted in the multidonor- supported Flood Action Plan, which started in 1990. The plan lasted five years and consisted of 26 components, varying from a wide range of studies to pilot projects. Although plan ended in 1995, a number of components continued with separate donor funding, such as the Compartmentalization Pilot project, financed by the Netherlands government. Substantial criticism was leveled at the FAP, especially in the early days when, mostly French engineers14 advocated the construction of massive embankments along all major rivers for "full" protection. These plans were not implemented, and FAP emphasis was shifted toward more realistic objectives. Nevertheless, it remained "fashionable to blame all the problems on the FAP" (Duyne 1998: 36). In retrospect, the program made a major contribution to the development of a more comprehensive approach toward water management in Bangladesh. Emphasis was now placed on social and environmental aspects as well as fisheries, navigation, and the need for full peoples' participation and 12 The Early Implementation Project, which ran up to the end of the 1990s, was set up following this shift in donor policy. 13 In 1988, more than 60 percent of the land in Bangladesh was inundated, and the crop loss was enormous. 14 Having its origin in the now famous visit to Dhaka by Danielle Mitterand, wife of the French president, when her plane was stranded at the airport during the 1988 floods. 24 Controlling or Living with Floods in Bangladesh consultation. The justification for large-scale public sector flood control, drainage, and irrigation projects came under increasingly stringent review, and the first comprehensive evaluation was made of the impact of FCD schemes in Bangladesh. Detailed guidelines for project, environmental impact, and social impact assessments were produced, and useful practical experience with specific technical options was gained through pilot projects, particularly on river bank protection, compartmentalization, flood platforms, fish passes, and dredging. Substantial further improvements were made in the country's database, particularly with spatial data, and a wealth of studies were made available in hardcopy form (Ministry of Water Resources 2001). Based on the outcomes of the Flood Action Plan, the government concluded that: [A] reformulation of the national water planning goals and objectives has now become necessary to guide future planning efforts and to produce an integrated national water plan (Ministry of Water Resources 1995 in Abdur Rahman 2002). This resulted the Bangladesh Water and Flood Management Strategy (1995, updated in 1997). The strategy became the working document for the water sector until the National Water Policy was made public in 1999, and subsequently a comprehensive National Water Management Plan (NWMP) was formulated (1999­2001). 4. Functions and Technological Elements of Flood Control and Drainage Systems Flood control, drainage, and irrigation are intricately connected in the humid and low-lying environment of Bangladesh. Drainage is not only about evacuating floodwaters and obtaining dry feet for agriculture and safe living conditions. The FCD infrastructure serves a multitude of water management functions and socioeconomic interests, sometimes mutually exclusive by nature. As a consequence, drainage in Bangladesh should be understood as water control in the broadest sense. Four characteristics stand out: · The conflict between different categories of land and water users. The Bangladesh National Conservation strategy identified six important areas of conflicting land uses: agriculture vs. shrimp and capture fisheries, forestland vs. shrimp and capture fisheries; agriculture vs. livestock; agriculture vs. settlements; agriculture vs. industries and brickfields, and agriculture vs. newly accreted charlands (Ministry of Environment and Forest 1990). · The complex management of water levels and quality in areas with competing and mutually exclusive demands on water levels and, in the coastal areas, salinity · The diversity of drainage conditions, even at the regional level · The variety of functions of FCD infrastructure such as protection of homesteads, agricultural land, and drainage (their primary functions), irrigation, navigation, transport over embankments, flood shelter, and housing. As a result, water management needs to · Integrate a multitude of types of water use--for agriculture, fisheries, and domestic use--and varying demands on timing, quantity and quality for each use · Consider diverse water and infrastructure functions, both consumptive and nonconsumptive such as navigation, transport over embankments and flood shelter. The foregoing points to many of the differences between FCD systems in Bangladesh and irrigation and drainage systems in arid or semi-arid environments. These differences are summarized in table 7. Physical Elements of Flood Control and Drainage Systems By excluding floods, the primary function of FCD infrastructure for agriculture is to convert F1, F2, or F3 land to shallower flood stages, which, if reliable, facilitates conversion from broadcast aman to transplanted aman and from local to more vulnerable high-yielding varieties. As a result, yields can increase. Additional functions include the protection of the boro rice crop from early flash floods in inland areas and the exclusion of saline water in coastal areas. Three types of drainage can be distinguished; · Gravity drainage under tidal conditions, practiced in the coastal zones where gravity drainage is possible only at low tide 25 26 Controlling or Living with Floods in Bangladesh · Gravity drainage under nontidal conditions, at the end of the monsoon season when water levels outside the embankments recede · Pumped drainage, on a limited scale, generally developed with the dual purpose of irrigation and drainage. Table 7 Differences between FCD and irrigation and drainage systems in (semi-) arid conditions Management aspect Irrigation and drainage systems in arid FCD systems in Bangladesh and semi-arid environments Infrastructure Infrastructure elements designed for a Infrastructure has to accommodate many, specific purpose (subsurface and surface often conflicting, demands. Demands drains, canals, structures) gradually change, increase and diversify. Design Manmade and designed for optimal Partly manmade and not designed for performance optimal performance Users Farmers with (relatively) homogeneous Many different users with heterogeneous demand; right water level or volume at requirements that are often contradictory and the right time mutually exclusive Operation Possible to plan in advance, use standard Difficult to plan in advance procedures Main management How to distribute water equitably in How to reach and implement a compromise challenge times of scarcity; waterlogging and for conflicting demands, including the salinity control? exclusion of particular uses of water or infrastructure? Financing Relatively easy to determine and Benefits differ considerably per user group, implement user financing, especially in are difficult to quantify, and can show large irrigation systems variation from one year to next Source: Adapted from Wester and Bron 1998. Classification of FCD Schemes Given the diversity of drainage conditions, developed technologies and the multipurpose nature of FCD systems, any classification is bound to hide local differences. A cut-and-dried distinction between drainage, irrigation, and flood control is therefore insufficient without an understanding of the prevailing land form to appreciate the particular nature of drainage in Bangladesh. A number of systems have been used to classify FCD systems, based mostly on the distinction between coastal and inland FCD and the availability of irrigation. The classification used by BWDB is presented in table 8.15 15In the National Water Management Plan, slightly different classifications are used Functions and Technological Elements of Flood Control and Drainage Systems 27 Table 8 Classification of flood control and drainage schemes Classification Description IRR Irrigation alone DR Drainage alone DI Drainage and irrigation together FCD Schemes involving only flood control and drainage and excluding coastal FCD. Can include irrigation inlets in embankments, to mitigate FCD effects on water flows and availability. FCDI FCD with measures for the supply of irrigation water, including khal excavation and with or without hydraulic structures CFCD Coastal FCD CFCDI Coastal FCD with measures for supply of irrigation water SCP Shrimp culture polder Source: Bangladesh Water Development Board 1998. A category also used is SFCD, partial FCD, with submersible embankments, occurring only in the Northeast (28 BDWB schemes) and North Central Regions (2 BDWB schemes). Here, these schemes are included as part of FCD and FCDI. The most common types of scheme are FCD, FCDI and CFCD, together constituting some 80 percent of all developed BDWB schemes. In table 9 an overview is presented of the schemes developed by the Bangladesh Water Development Board. Table 9 Overview of scheme types and sizes Scheme size Type of scheme Small Medium Large Very large Total 0­1,000 ha 1,000­5,000 5,000­ >15,000 ha ha 15,000 ha Irrigation (IRR) Number of schemes 22 10 n.a. 2 34 Area in ha 10,614 23,266 n.a. 335,084 368,964 Drainage (DR) Number of schemes n.a. 19 9 9 37 Area in ha n.a. 48,308 80,257 331,698 460,263 Drainage and Irrigation (DI) Number of schemes 10 28 12 2 52 Area in ha 6,317 66,902 90,349 43,556 207,124 Flood control and drainage (FCD) Number of schemes 7 42 41 25 115 Area in ha 2,473 161,895 383,294 1,469,185 2,016,847 Flood control, drainage and irrigation (FCDI) Number of schemes 13 56 49 25 143 Area in ha 6,929 161,895 404,183 1,356,687 1,929,694 Coastal flood control drainage (CFCD) Number of schemes 5 57 42 19 123 28 Controlling or Living with Floods in Bangladesh Table 9 Overview of scheme types and sizes Scheme size Type of scheme Small Medium Large Very large Total 0­1,000 ha 1,000­5,000 5,000­ >15,000 ha ha 15,000 ha Area in ha 3,157 176,284 332,336 443,828 955,605 Coastal FCD and irrigation (CFDI) Number of schemes n.a. n.a. n.a. 2 2 Area in ha n.a. n.a. n.a. 163,880 163,880 Shrimp culture polder (SCP) Number of schemse 5 n.a. n.a. n.a. 5 Area in ha 2,387 n.a. n.a. n.a. 2,387 Town protection schemes (TP) n.a. n.a. n.a. n.a. 76 Total Number of schemes 62 212 153 84 587 Area in ha 31,877 638,550 1,290,419 4,143,918 6,104,764 Distribution (percent) Number of schemes 12.1 41.5 29.9 16.4 100 Area in ha 0.5 10.5 21.1 67.9 100 n.a. Not available. Regionally, the distribution is as shown in table 10. Source: BWDB 1998 Table 10 Analysis of BDWB implemented schemes Hydrological region CFCD CFDI DI DR FCD FCDI IRR SCP TP Total Central 0 0 6 10 30 35 6 0 17 104 Northeastern 0 0 34 3 25 33 5 0 15 115 Northern 0 0 0 2 14 16 9 0 9 50 Northwestern 0 0 2 3 27 11 0 0 6 49 Southeastern 35 1 4 0 1 14 8 5 15 83 Southern 46 1 0 4 5 10 1 0 7 74 Southwestern 42 0 6 15 13 24 5 0 7 112 Total 123 2 52 37 115 143 34 5 76 587 Source: BWDB 1998. Components of FCD Systems In this report, flood control and drainage systems are characterized as: . . . all areas in the flood plains of the rivers in Bangladesh and in the coastal plains that are utilized by people, and that contain some or all of the following infrastructure: Functions and Technological Elements of Flood Control and Drainage Systems 29 embankments, regulators, khals, beels, cross bundhs/dams and canals (Wester and Bron, ILRI 1998: 11). After first summarizing their main water management functions, each physical component of the FCD systems is discussed in detail. · Embankments protect the area from (early) flooding. · Regulators within an embanked area serve to drain and retain rainwater. Vertical lift gates, fallboards and flap gates can be an element of regulator structures. · Khals are natural drainage channels both inside and outside embanked areas. · Beels are naturally occurring (semi-)permanent water bodies, connected to the river system by khals. · Haors are a separate and unique type of deep-flooding water body found only in the Northeast region · Pump stations, found only in a small number of FCDI systems, have a dual drainage and irrigation purpose. · Pipes and culverts, small structures usually installed by users, serve local (not system) drainage and irrigation functions. Embankments. Embankments are the single common feature of all FCD systems. (see photo) Their primary function is to protect polders or areas from river floods or tidal action. This enables the cultivation of flood- sensitive crops such as high-yielding transplanted aman in monsoon season and protects aus and boro rice crops from flash floods in the pre- and post- monsoon season. An additional function, which is nearly universally appreciated, is the protection of homesteads, cattle, roads, and buildings. Traditionally, embankments have been designed as single-purpose Embankment and road infrastructure, whereby the dwelling William J. Oliemans function (which is formally illegal) has been strongly discouraged. Embankments are designed with a free board of 3 feet above the flood level or tide and with a 20-year return period (table 11). If certain embankments are turned over to user or local government organizations, as foreseen in the National Water Policy, the (uniform) use of 20-year return periods may cause a false sense of security, posing additional risk to life and property. As a comparison, in the Netherlands, different return periods are used, depending on the value of the property protected, the 30 Controlling or Living with Floods in Bangladesh "prediction and response time" (for emergency evacuation), and the potential for damage.16 In Bangladesh, sea-facing embankments are typically constructed with a 1:7 seaside slope; river-facing embankments, with a 1:3 or 1:5 slope. Submersible Embankments. Located in the Northeast in the haor area, these embankments serve to delay the onset of the monsoon floods and protect the standing rice crop against flash floods. Traditional management is a main feature of these embankments, and the Bangladesh Water Development Board has been much less involved in these areas than in the rest of the country. Embankments of this kind are constructed and maintained primarily through community action, sometimes involving massive organizational efforts, in which flood levels and the condition and height of the embankment are monitored day and night. Reasons for the extensive community involvement in the haor areas include the need for collective action (there is no time to wait for government projects!) and the isolated location, especially in the monsoon season when most of the area is under water. Some 28 submersible embankments have been constructed by the BWDB in the Northeast, most of them in Mymensingh District (Ministry of Water Resources 2001). Table 11 Bangladesh Water Development Board standard embankments designs Side slope Minimum set- Type of Country- Riverside Crest width Minimum back from Return period embankment side and seaside (m) freeboard (m) bank (m) (years) Sea 1:2 1:7 4.2 1.5 76 20 Interior (river) 1:2 1:3 or 1:5 4.2 0.9 53 20 Marginal 1:2 1:2 2.4 0.9 38 20 Source: Md. Liakat Ali 2002. Drawbacks of Embankments. Despite widespread appreciation of the protection function provided by embankments, serious drawbacks have also been identified. They include: · Drainage congestion and waterlogging, with embankments keeping floodwaters out and rainwater in. Insufficient drainage outlets were provided, especially on the initial projects, leading to widespread drainage congestion. As a consequence, area inhabitants of the area cut the embankments. · Interruption of fish migration routes. Bangladesh's inland fisheries are intimately bound to the sequence of annual flooding. The monsoon flood connects rivers, tributaries, and all static water bodies to a dynamic flood plain system. As the plains become inundated by rainfall and riverine flooding, many species begin a longitudinal migration upstream, often to spawn. Once the rivers begin to flood, there is a lateral migration into distributory channels and flood plains that provide a rich feeding ground for hatchlings and fry for four to five months. At the end of monsoon when floodwaters recede, the expanded fish population starts moving back into the rivers, beels, and other permanent water bodies, which are then used for fishing (World Bank 1998). Embankments have typically cut across these migration routes, which, together with the gradual decline of the dry-season water bodies, has had a negative effect on capture fisheries. According to monitoring data from the Department of Fisheries, between 1983 and 1989, from a base of nearly 500,000 t/year, fish production from rivers, flood plains, and beels has declined by 44,000 t/year. 16For sea dikes, 1:10,000 years; for large river dikes, 1:1,250 years (the norm for "state waters"); for internal drainage dikes, 1:100 years; and for minor embankments, less than 1:10 years Functions and Technological Elements of Flood Control and Drainage Systems 31 In the same period however, the inland aquaculture (or culture fisheries) output rose with 48,000 tons a year17, making up for the loss of inland capture fisheries. Reportedly the situation has stabilized. Other factors contributing to the reported decline in capture fisheries include: the reduction of fish breeding and nursing habitats due to the decline in dry-season water bodies, the massive capture of fry for aquaculture, increased fishing intensity, increased use of agrochemicals, and diseases attacking especially beel fish (Compartimentalization Pilot project 1999). · Raised flood levels outside the embanked areas. Data analysis by Anjan Datta in 1999 points to a 35 percent increase in rice production in FCD schemes (Md. Liakat Ali 2002), and the positive impact on agricultural production is widely recognized. At the same time, Flood Action Plan studies indicate that, due to confinement of the drainage channels, flood levels outside the embanked areas have increased. Although no systematic monitoring has been undertaken, a 1998 study by the Bangladesh University of Engineering and Technology in 20 thanas showed no significant correlation between the percentage increase in FCD coverage and the percentage increase in aman production per thana between 1964 and 1993. The negative external impacts such as heightened flood levels and disruption of drainage channels appear to largely cancel out the positive internal impact (BUET 1998 in Ministry of Water Resources 2001). · The lack of maintenance. Late or no maintenance has been a problem faced in nearly all areas. Recently, under the Coastal Embankment Rehabilitation Program in the Southeast, embankment settler groups were formed. In exchange for a minor subsidy and a homestead on the land-facing embankment, these groups have been made responsible for maintaining the embankment and its road. In the Dampara Water Management project in the Northeast, the inhabitants were encouraged to plant vetiver grass as an income-generating activity, which also gave them an incentive to maintain their embankments. Though perhaps too soon to tell, these initiatives may provide a way out of the maintenance trap common to so many of the country's embankments. The absence of a cost-recovery mechanism has contributed to this problem. · Lengthy land acquisition procedures. Another constraint is posed by lengthy procedures for acquiring land for embankments. Land is acquired by the central government and leased to the Bangladesh Water Development Board. The board, in turn, pays a substantial annual tax to the Ministry of Land. Delays in project implementation, a constraint faced by many FCD projects, are often incurred in the process of land acquisition. A participatory process, as applied recently by BWDV in the Dampara Water Management project (funded by the Canadian International Development Agency), suggests that by involving stakeholders in defining the alignment of the embankment, the acquisition process can be greatly shortened (Bangladesh Water Development Board 2001). · Sedimentation of (large) drainage channels. Interference with tidal flooding of the land by the construction of embankments can lead to rapid silting up of the adjoining rivers and creeks (Brammer 1996). This is due to the slow withdrawal of stagnant water from the channels after high tide, which gives sedimentation more time to settle than in natural conditions. The development of coastal polders in the Southern regions of the country has channeled rivers between polder embankments and the resulting sedimentation has in many areas caused prolonged drainage congestion. In severe cases, 17Output is still growing at a rate of nearly 10 percent a year (Ministry of Water Resources 2001). 32 Controlling or Living with Floods in Bangladesh former agricultural communities have had to reincorporate fishing into their income-earning activities. The case of Dakatia Beel illustrates this situation .18 · Proposed solutions to this serious problem have been dredging and the development of rotational tidal basins, to increase the tidal flow and enable sedimentation in the basin. Given the land pressure and socioeconomic conditions in the area, the social impacts of applying this approach on a wider scale should not be overlooked. · Interrupted navigation routes. Although traditional boat navigation routes have been interrupted, the benefits of improved road transportation have typically far outweighed this negative impact. Embankment Cutting. A widespread feature of embankments is cutting, also referred to as public cuts. Cuts take place for various reasons: · To allow congested flood waters, usually the result of intense rainfall, to drain from the embanked area into the rivers. · To allow fish migration in the early monsoon season (haor areas). · To allow navigation in the early monsoon season (haor areas). · To provide relief to areas outside the embanked areas, typically by persons outside the embanked area who consider that the embankment causes higher flood levels due to the decreased storage. · To allow saline water entry for salt and shrimp production (Southwest, Central, and East). · To allow tidal irrigation. Many of these cuts are planned water management measures taken by users, and not the result of ignorance or vandalism. After comparing the costs of embankment breaching and repairing with the cost of operation and maintenance (O&M) of sluice gates, the National Water Management Plan concluded that "over 25 years, vents cost on average 2.2 times as much as the annual cost of breaches." At the onset of the floods, the embankments are often repaired by the very same individuals who opened the breach (Duyne 1998). Regulators. Regulators, both major and minor, allow for the drainage of rainwater from the drainage channels into the river system and for the controlled passage of river water through the system. · Major regulators are constructed at the head of a large khal at the same time as the embankment. · Minor regulators are either constructed simultaneously on minor khals or added later during remodeling or rehabilitation of the system. Regulators are equipped with vertical lift gates, flap gates, or both at the riverside and fall boards at both the countryside and the riverside. The lift gates and fall boards allow maintenance of the gates and water retention on the countryside for fishing and agricultural purposes. 18"Waterlogging again in Beel Dakatia and Bhabadaha," Bangladesh Observer, July 21 2002, Functions and Technological Elements of Flood Control and Drainage Systems 33 An additional, essential function of regulators is to maintain or increase water levels for irrigation, fisheries, and domestic purposes. The inlet of water for irrigation is referred to as flushing in Bangladesh. Management arrangements for operating these structures have been highly developed. Services are auctioned or tendered to small private contractors (often local farmers) who take care of the inlet of irrigation water from outside the embankment, distributing water, and collecting the fees. Regulators are the main structure for managing water levels in an area and as such they are the focal point for management interventions and, subsequently, conflicts. Typically, fisheries favor water retention, as do farmers farther inland from the regulator or highland farmers wishing to irrigate a boro rice crop. If regulators drain a large area, they do not permit close control of water levels, often precipitating conflicts and prompting individuals or groups to take action on their own. Typical actions include opening, closing, or breaking the gates and constructing cross-dams in khals upstream of the regulators to allow for local water level control (Wester and Bron 1998). The potential benefits and conflicts arising from the operation of regulators are summarized in table 12. Khals. The natural drainage channels in Bangladesh are referred to as khals. A common feature in all areas, khals are used for navigation, fisheries (migration and capture), water retention, and drainage. In many cases, the pattern of minor and major khals was remodeled, typically by closing some minor khals and allowing one or two major khals to take over their drainage function. In practice, however, drainage congestion recurs in localized areas, and additional structures are required to relieve local drainage problems. Such structures are typically small, private, drainage pipes serving a small area and do not usually cause conflicts. In FCD systems, drainage has been the main design function of khals, but their actual functions vary over time. During the monsoon, when drainage needs are highest, drainage is their primary function. During the dry season, water retention is their main function, especially for boro rice cultivation and fisheries. Table 12 Gate operation: benefits and conflicts Gate operation Potential benefits Possible conflicts of interest Opening gates Allows accumulated rain water to drain out Highland farmers want to store water premonsoon for of system, preventing damage to early aus for irrigation or land preparation. drainage and late boro on low land. Allows entry of Saline water is detrimental to paddy saline water for salt and shrimp production production. or for fish culture. Closing gates during Prevents monsoon river flood damage to Prevents fish from entering into monsoon to prevent aus and aman crops. system. Accumulated rain water flooding from river flood system. Opening gates for Allows entry of water with fish fingerlings. Extended flooding damages aman flushing during Supplies extra water for aman on higher rice on low land. monsoon land. Closing gates post- Retains water for flowering of aman crop Reduces fish catch in khals and monsoon for water and for irrigation of boro crop. Retains beels. Makes boro rice cultivation in retention water for domestic purposes beels difficult. Opening gates post- Increases fish catch in khals and beels Highland farmers want water. monsoon for drainage Drains beels for boro rice cultivation. Retention for irrigation. Reduces amount of water retained for domestic purposes Source: Adapted from Wester and Bron 1998. 34 Controlling or Living with Floods in Bangladesh The main management interventions in the khals are reexcavation and the construction of cross-dams (bundhs). Cross-dams are semipermanent (see photo) structures erected to store or divert water for irrigation, fisheries, or both. They are typically built of earth, stones, and other locally available materials. Excavation is needed to maintain the drainage and storage capacity of the khals, while cross-dams are used to raise the water level and store water for irrigation and for fisheries. As less water enters the area, the drainage capacity of the khals becomes larger than needed after embankment construction. Nevertheless, drainage congestion still occurs due to the construction of cross-dams and development of fish ponds in the khal. In the dry season, this does not cause any major congestion problems, but at the onset and end of the monsoon Cross-dam erected in a khal season these structures reduce drainage Source: William J. Oliemans. capacity and cause flooding and crop losses. The different functions of khals are summarized in table 13. Table 13 Function of khals in different seasons Season Use Monsoon Post-monsoon Dry season Drainage ***** *** * Navigation **** *** n.a. Fishing **** **** ** Water retention for irrigation n.a. **** ***** Water retention for fishing n.a. *** **** Tidal irrigation n.a. **** ** Entry of saline water for salt and **** **** **** shrimp production (Coastal areas) Domestic purposes ** *** ***** Livestock watering and washing ** *** **** Boro rice seedbeds n.a. **** n.a. Boro rice cultivation n.a. ** ***** Note: * indicates relative importance. Source: Adapted from Wester and Bron 1998 Beels. Beels are naturally occurring (semi-)permanent water bodies that fill up and expand into huge lakes in monsoon season and decrease or dry up in the dry season. Beels can be found across the country except in the coastal areas and the Eastern Hills. Beels are connected to river systems and other beels through khals. Beels mainly serve fisheries, but their other functions include water storage for irrigation, domestic purposes, and navigation. As the beel level recedes, the moist soil exposed is used to cultivate boro rice and some other dry season crops. The differing demands of capture fisheries and agriculture cause major conflicts surrounding beels. Maintaining the highest possible water level is in the interest of fishermen, but farmers in the vicinity Functions and Technological Elements of Flood Control and Drainage Systems 35 want to drain the beel to start growing their rice. Farmers located farther away and on higher land see the beel as a storage reservoir for irrigation water, which poses other, contrasting demands. Haors. The haors are a unique water body found only in the Northeast region of Bangladesh in the Sylhet Basin (see photo). This area, which has been down-warped due to tectonic influences (Brammer 1996),19 is a huge (5,600 km2) depression with areas that flood up to a depth of 6 m in monsoon season. The Haor Basin consists of some 47 major haors and 6,300 beels, 3,500 of them permanent, the remainder seasonal. The haor area knows only two seasons, monsoon, Deep flooding in a haor area used primarily for large-scale William J. Oliemans fishing, and the rabi season, used for cultivating boro rice. An ingenious system of contour bunds has been developed to allow for flood-recession agriculture. In addition, a system of canals, built and maintained by rice growers, supplies the areas with irrigation water. Water is lifted to the fields with swing buckets, and, since the introduction of mechanized irrigation in the 1970s, with low-lift pumps. Conflicts are fewer than in the beel areas. The haors are mainly fishing areas and, unlike the other inland systems, agriculture is not usually the only household occupation. Boro rice is the only crop.20 Conflicts arise between fishermen and farmers about timing the embankment breach in the early monsoon. Fishermen typically favor an early breach to allow fish migration to the flood plains and entry of fishing boats; farmers favor a delayed breach to allow the rice crop to ripen. Flapgates. The main function of flapgates is to allow for automatic drainage of excess water from partially or fully embanked areas or polders. Difference in water level between the country-polder side and the river- and seaside cause flapgates to open automatically and close as soon as the outside water level rises. Flapgates are double-hinged to limit pressure on the hinges from frequent blockages. To flush (irrigate), flapgates should be held open. As this important function is normally not incorporated in the design, the gates are often damaged in attempts to keep them open for irrigation. Pump Stations. Only a few FCDI systems21 have pump stations, and most of them serve a dual purpose of drainage in the monsoon season and irrigation in the dry kharif and rabi seasons (for aus and boro rice). Serious doubts have been raised about the economic viability of pumping, and it is generally not considered an option (Ministry of Water Resources 2001). 19The area is said to be continuing to subside, at a rate of 1 cm a year (Environmental and GIS Support Project for Water Sector Planning, Ministry of Water Resources 1998). 20 Farmers have been known to harvest the fully submerged rice crop by diving under water 21Such as the Chandpur and the Meghna Donagoda Irrigation project in the Southeastern region. 36 Controlling or Living with Floods in Bangladesh Development of Flood Control and Drainage Systems Safwat Abdel-Dayem (2000) classifies drainage systems according to the primary function and the climatic zone in which they occur. (Accordingly, Bangladesh, located in the humid tropical climate zone, can be characterized by its primary function of flood control and, to a lesser extent, by land reclamation and irrigation. Evolution of FCD Systems in Humid Tropics Flood control systems in the humid tropics generally start with the construction of embankments to keep out seasonal river or tidal flows. Over time, flood control systems can become increasingly complex and sophisticated with, for instance, the construction of adjustable gates in the embankments to release (or take in) water. As control over water increases, farming intensities may also increase. With intensified farming practices, more water may have to be retained within the system during the dry season, putting additional demands on water control technology. Development Phases in FCD Systems in Bangladesh A similar picture can be painted for Figure 6 Unprotected flood plains Bangladesh. Four FCD development phases can be distinguished: unprotected flood plains, flood protection, reduction of drainage protection, and optimized water control (Wester and Bron 1998). Phase 1--Unprotected Flood plains (fig 6). Flood control and drainage practices exist without government intervention. Individuals control water through small embankments, cross-dams, and drainage canals. Flood plains are subject to seasonal flooding to which cropping patterns and varieties are adapted. Yields are generally Illustration: Djouke Joosten. By permission (Wester and Bron 1998). low, and traditional irrigation takes place by lifting beams or swing buckets. Figure 7 Flood protection Phase 2--Flood Protection (fig 7). In this phase the main objective is to increase crop security. Government interventions are geared toward controlling floods from the river and sea for which embankments are built. The same embankments cause drainage congestion from accumulated rainwater within the protected area or from runoff from higher land bordering the area. Usually, drainage openings are created in the embankment, equipped with flap gates on the riverside and regulators on the main outlets. At the end of the monsoon season, Illustration: Djouke Joosten. By permission (Wester and Bron 1998). drainage becomes more congested than Functions and Technological Elements of Flood Control and Drainage Systems 37 before the embankment was constructed. Floodwater evacuation is now confined to the main drainage arteries equipped with regulators, and the area does not drain naturally as the river water level falls through the khals network. People intervene by cutting embankments to relieve the congestion and, at the onset of the next monsoon, fill in these cuts. Coastal polder systems consist of a ring dike with one main canal and draining outlet, which means that only one water level can be regulated. Local differences in water level to accommodate the needs of different producers and sectors (fishermen, rice growers, jute growers) are regulated through cross-dams. Phase 3--Reduction of Drainage Congestion Figure 8 Reduction of drainage congestion (fig 8). This phase deals with alleviating congestion caused by the construction of embankments, gates, and regulators. A second round of engineering interventions takes place, and minor regulators are installed on the smaller drainage channels, additional surface drainage outlets are constructed; khals are excavated regularly. Water levels can be managed in smaller units, and this phase marks the optimization of flood control and drainage for crop production. Djouke Joosten. By permission (Wester and Bron 1998). Figure 9 Optimized water control Phase 4--Optimized Water Control (fig 9). With lower flood risks, farmers invest in high-yielding varieties, and the cropping pattern expands. An additional crop can be grown. Crops are now grown into the dry period, raising demand for water in the dry season. Users devise means to retain water within the system, for example, by opening flap gates. Another remodeling round takes place, now to facilitate retention or inlet of water. Lift gates are installed, khals are deepened and widened, and control structures are installed within the system. A substantial part of the area is now irrigated in the dry season by traditional or Illustration: Djouke Joosten. By permission (Wester and Bron 1998). mechanized irrigation from surface water and groundwater. In coastal areas, water management becomes more sophisticated, by equipping the flap gates to allow regulated inflow and to manage salinity. Typology of Drainage Issues and Systems In defining the main drainage issues in Bangladesh, three levels of analysis have been used: the components of FCD systems; the hydrological region; and the land form or FCD system. The hydrological region is above all a planning and institutional organization, geared toward macro-level resource management, to steer policy and allocate resources at the regional level. Analysis at land form­ FCD system level facilitates the understanding of the interaction between natural conditions and 38 Controlling or Living with Floods in Bangladesh manmade interventions. Given the complexity of FCD systems as described, this typology facilitates planning of physical improvements and development of institutions at the local or subregional levels. Hydrological Regions In the National Water Policy and the National Water Management Plan, water resource and portfolio planning is based on eight hydrological regions, each representing a specific catchment area (except for the Rivers and Estuaries region). Table 15 summarizes the main drainage and environmental issues in these regions. Main Land Forms--FCD Systems The Systems Rehabilitation project in 1997­98 developed a classification of FCD systems based on an in- depth water management appraisal of 27 FCD systems spread over Bangladesh. Instead of being a classification of FCD systems only, a combination of land forms and FCD systems has been described, a methodology that facilitates the analysis of the interaction between the FCD systems and their biophysical and socioeconomic environment. Land forms are the result of human interventions in the natural environment through both formal "rehabilitation" or "remodeling" and as result of "informal" adaptations by users. The main defining characteristics of land forms are: the type of FCD infrastructure, the nature of the floods, interactions with the flood plains, and geomorphology. The main drainage characteristics of the land forms are summarized in table 14. The main characteristics of the land forms are summarized in table 16 and depicted in figure 10. Table 14 Main characteristics of the land forms, FCD systems Land form- Geomorphic area Flood type Type of drainage infrastructure Hydrological FCD system region Deltaic coastal Ganges-Meghna Tidal, river § River-andsea-facingembankments Southwest, polder deltaic flood plain § Interconnectedkhals Central, and (Khulna-- § Majorandminorregulators East Noakhali) § Flapgates,sluicesandpipes Nondeltaic Chittagong/Meghn Rainfall, § Sea(river)facingembankmentswith Eastern Hills coastal Polder coastal plain flash parallel khals perpendicular to the (Chittagong) embankment § Flapgatesandregulators Beel Pleistocene Rainfall, § Onemainriverembankment, Northwest, terraces, flood river regulators, khal closure and North Central plains, Atrai Basin excavation § Connected beels, a few large and many small Flood plain Inland flood plains River, § One or two main river embankments Across country of the major rivers rainfall in network of rivers except Eastern and tributaries § Manykhalsinarea Hills and § Beels,notconnected,mainwater Northeast bodies Hoar Sylhet Basin River, § Submersibleembankments,haors, Northeast rainfall, beels and khals flash Source: Adapted from Wester and Bron 1998. Table 15 Hydrological regions: main environmental and drainage characteristics Gross area Average Main environmental concerns Type/cause of Proposed drainage interventions Region (km2) rainfall (priority concerns, NWMP) flooding/drainage congestion (NWMP) (mm) Southwest (SW) 26,226 1,655 Low river flows, including salinity; Tides and cyclonic rainfall, Dredging; diversion of river flow; sensitive areas; safe domestic decrease of dry season local area management with water supply; surface water Ganges flow after Farakka participation and consensus quality; sedimentation and dam construction. navigation Sedimentation of river channels South Central 15,436 2,307 Low river flows; safe domestic Tides, cyclonic rainfall and Dredging, diversion of river flow (SC) water supply; cyclones; surface surges, overspill of Lower water quality; sensitive area Meghna 40 management; arsenic Northwest 31,606 1,739 River bank erosion, surface water Locally intense rainfall, Dredging outlet Atrai to the (NW) quality, low rainfall, decline of the impeded drainage, breaches in Brahmaputra; improve existing water table and permanent water Teesta and Brahmaputra Right drainage structures. Improve bodies, decline of fisheries Embankments, breaches connectivity FCDI­flood plains- resources in internal polder water bodies. Embankment embankments, drainage construction with adequate drainage congestion provisions due to high river flows North Central 15,949 1,956 Main river flooding, rainfall Locally intense rainfall, Urban drainage, dredging. Improve (NC) flooding and drainage congestion, impeded drainage, spillage connectivity FCDI-flood plains- main river bank erosion, from the Brahmaputra and water bodies. Embankment groundwater availability, fisheries congested drainage construction with adequate drainage resources on Meghna provisions Northeast (NE) 20,061 3,194 Fisheries, flooding and drainage, Flash floods on transboundary Submersible embankments, flood safe domestic water supply, rivers, locally proofing, urban drainage, sensitive area management, intense rainfall, impeded embankment construction with Table 15 Hydrological regions: main environmental and drainage characteristics Gross area Average Main environmental concerns Type/cause of Proposed drainage interventions Region (km2) rainfall (priority concerns, NWMP) flooding/drainage congestion (NWMP) (mm) sedimentation drainage and drainage adequate drainage provisions congestion on the Meghna Southeast (SE) 10,284 2,271 Safe domestic water supply; See NE, including tidal Embankments, flood proofing, flood surface water quality; rainfall influences warning, remodel drainage network flooding; drainage; cyclones; to solve problems with coastal cross- arsenic dams, remodel and properly operate regulators, urban drainage. Improve connectivity FCDI ­flood plains- water bodies Eastern Hills 19,956 2,445 Coastal zone: cyclones, safe Flash floods and cyclonic Zoning, land registration, 41 (EH) domestic water supply, surface rainfall embankment construction if land has water pollution, water management reached equilibrium, improved and social conflicts. regulator design and operation Hills region: safe potable water supply, deforestation, ,water- related diseases, sensitive area management, erosion Rivers and 8,607 2,318 Erosion, accretion, safe domestic River floods, tidal floods Package of flood proofing, Estuaries (RE) water supply, cyclones (estuary), preparedness & response, navigation embankment protection, construction & retirement, reclamation and settlement programmes to include effective drainage and user participation Source: Ministry of Water Resources 2001, 2002. Functions and Technological Elements of Flood Control and Drainage Systems 41 Figure 10 Main land forms, flood control and drainage systems Deltaic coastal polder (khulna, noakhali) Nondeltaic coastal polder (Chittagong) Beel Flood plain Haor Illustration: Djouke Joosten. Source: By permission (Wester and Bron 1998). Table 16a Characteristics of the different FCD systems Coastal deltaic polders FCD system Coastal Nondeltaic polders (EH) Beel (NW, NC) (SW, SC, SE) Main produce Both T. aus and T. aman and where possible a Mostly rice in the Northern Coastal Diversity of cropping; rice, banana, Rabi crop, shrimp and limited salt, fish Plains (Chittagong) and jute, fruits, vegetables, pulses Salt/shrimp/rice in the Southern Coastal Plain (Cox's Bazar) Issues and challenges in Drainage and Irrigation under tidal conditions Controlling salinity Monsoon flooding by rainfall/runoff water management Control of water levels inside the polder for Erosion of embankments (sea/river). from higher areas. different needs. Management of drainage under tidal Balance beel preservation for fisheries Controlling salinity. influence with cyclones/tidal surge with drainage for agriculture. Balancing interests of shrimp and rice Balance interests agriculture­salt Maintain fish migration routes: Maintaining drainage capacity outside polders; production­shrimp growing. development of fish-friendly siltation reduces drainage base. Control of Flash floods from the hills structures Allowing sufficient accretion in char lands before Distribution of irrigation water along Drainage constrained due to high 43 settlement. streams and creeks (upstream-- river water levels and silting up khals downstream). Dry season: Drought and falling groundwater table means shallow tubewells may go out of production Loss of soil fertility and impeded evacuation of polluted water. Type of infrastructure River/sea facing embankments Sea (river) facing embankments with 1 main river embankment, regulators, and intervention Many interconnected khals, most of which khals running in parallel through the khal closure and excavation closed\realigned and where waterlogging again embanked areas, perpendicular to the Many connected beels, a few large takes place embankment and draining both and many small Many minor regulators, flapgates, sluices and monsoon rains and floods. In the "newer" systems: flushing pipes. Embankments with flapgates and inlets, tubewells and low-lift pumps low-lift pumps, typically few tubewells. regulators. on surface canals (khals, irrigation Experiments with tidal basins for sedimentation Cross-dams constructed in creeks and canals). and tidal flow minor rivers for irrigation purposes Low lying parts characterized by a Macro level: dredging of rivers to maintain In shrimp areas additional inlet system of connected beels (some drainage function structures developed by shrimp perennial); large and small. cultivators Cross-dams in creeks/rivers for irrigation Table 16a Characteristics of the different FCD systems Coastal deltaic polders FCD system Coastal Nondeltaic polders (EH) Beel (NW, NC) (SW, SC, SE) Conflicts Shrimp -rice growers Shrimp/salt-rice in southern coastal Agriculture--fisheries (beel Inside/outside polders zone leaseholders) Low-high; large-small distance to inlet Agriculture-fisheries in Highland--lowland farmers, northern/middle zone Navigation Upstream/downstream use of surfacewater for irrigation Livelihood strategies Move from fish capture to agriculture. Diverse labor opportunities fisheries, Agriculture with intensive boro. Capture/culture fisheries remains important. industries, construction, transport, Increase of safety, mobility and Freshwater shrimp emerging. migration. alternative employment. Fishermen/ boatmen seek alternative employment (trade, brickfields, construction, etc). Recommended Enhancement of legal status user organizations. Less experience with user Basin level planning and interventions Piloting with Water Management Board. organization than other areas, need for development, better linkages LGI-- 44 Maintaining of drainage base outside the polders. capacity building, see Coastal Deltaic BWDB. Preservation of dry season Zoning shrimp--nonshrimp Polders water bodies, improving connectivity (fisheries, biodiversity). Pilot with Water Management Board model Table 16b Characteristics of the different FCD systems FCD system Flood plain (NW, NC, SW, SC, SE) Haor (NE) Main produce Great diversity of cropping; rice, sugarcane, banana, jute, Predominantly rice--boro mango, other fruit, potatoes, vegetables, pulses and spices Issues and challenges in water Controlling river floods Flash floods in the premonsoon threaten standing management Obstruction of drainage due to insufficient drainage outlets in boro crop embankments, roads, and lack of excavation of khals. Maintenance submersible or dwarf embankments Decline of soil fertility. Complex system of retention and flooding Maintain fish migration routes: fish-friendly structures Timing of the cutting of the embankment to balance Maintaining dry season water bodies for fish the interests of fishermen, boatmen and agriculture Confinement caused by embankment increases flooding in unprotected areas Decline GW table Table 16b Characteristics of the different FCD systems FCD system Flood plain (NW, NC, SW, SC, SE) Haor (NE) Type of infrastructure and intervention As beel system but 1 or 2 main river embankments in the middle Haor: large--and deep-water body flooded 6 to 8 of a network of river months a year, with a small number of beels Many khals and small rivers in the area Cutting of embankment a conscious management Beels (not connected as in Beel system) are the main water tool bodies In post-monsoon construction of a series of bunds Most shallow tubewells and deep tubewells located in this area for water retention or `compartments' Irrigation through STW/DTW negligible. Conflicts Agriculture--fisheries Large-scale entrepreneurial fisheries--agriculture Inside--outside embankment Timing of letting in floodwaters for fish vs. harvesting boro crop. Livelihood strategies Move toward highly intensive cropping (in some areas 3 crops), Isolated in the monsoon; poor communication, increased and more secure agricultural production. Decline in nutrition and health. User organization need to cope 45 capture fisheries. with crisis situations. Compact settlements in high areas. Migration in monsoon to collect boulders (Barki) and fisheries. Recommended interventions See beel See beels, build on local institutions, maintain linkages haors--river systems Functions and Technological Elements of Flood Control and Drainage Systems 45 Effects of Projected Changes in Climate Global warming is expected to have especially severe effects on Bangladesh and exacerbate many of the challenges discussed in tables 14 and 16.The most recent projections for Bangladesh include sea level rise, increased monsoon rains, and an increase in air temperature in the Bay of Bengal (table 17). Table 17 Effects of global warming on tides Year Sea level rise Temperature increase (°C) Change in rainfall compared to 1990 (cm) 2030 30 +0.7 in monsoon; +1.3 in winter ­3 percent in winter; +11 percent in monsoon ­37 percent in winter; +28 percent in 2050 50 +1.1 in monsoon; +1.8 in winter monsoon Source: World Bank 2002. Climate change will also affect flows in the transboundary rivers. Temperature changes will affect the timing and rate of snow melt in the upper Himalayan reaches, which would alter the flow regime in the rivers that rise in the Himalayas. Lower dry season rainfall and increased water demand due to higher temperatures will increase abstractions from rivers upstream and reduce the flow reaching Bangladesh (Ministry of Water Resources 2001). Consequences of Climate Change The key consequences of climate change for drainage in Bangladesh are: · Increased flooding due to increased monsoon season rainfall. For example, a 10 percent increase in precipitation may increase runoff depth by one-fifth and the probability of an extremely wet year by 700 percent (World Bank 2002). · Worsened drainage congestion, waterlogging, and flooding due to higher sea levels and rise in river bed levels. The continuous development of infrastructure (e.g. roads) will aggravate the problem and further reduce the limited natural drainage capacity. One of the key effects of drainage congestion is that it will increase the period of inundation and will expand wetland areas.22 · Lengthened tide-locked period. Higher sea levels will lengthen the tide-locked period for tidal drainage sluices and reduce their drainage capacity. According to the Intergovernmental Panel on Climate Change (IPCC), the rise in sea level would be in the range of 15 cm to 95 cm by 2100. Even a cautious projection of a 10 cm sea level rise, which would most likely happen well before 2030, would inundate 2,500 km2, 2 percent of the total land area. The areas around Patuakhali, Khulna, and Barisal in the Southwest and South Central regions, are most at risk. On average, the sea would move in about 10 km, but in the Khulna region, the sea will likely move farther in. With the high-end estimates, sea level rise in Bangladesh would inundate 18 percent of the country by 2100. 22There may be some benefits for fisheries and for boro production 46 Controlling or Living with Floods in Bangladesh · Irreversible damage to biodiversity. Increased salinity may prevent reversal of damage to the biodiversity of natural reserves such as the Sundarbans. Measures proposed to combat the expected increase in drainage congestion include: 1. Increasing the drainage capacity of infrastructure 2. Installing new regulators, using adjusted design criteria 3. Developing tidal basins 4. Installing pumps for drainage 5. Filling land, using natural or artificial methods to reduce inundation and promote drainage.23 Increasing the drainage capacity of existing infrastructure such as roads seems a feasible and effective way of reducing drainage congestion where drainage is hampered by culverts, bridges, regulators, and the like. New regulators and tidal basins are alternative physical interventions. Tidal basins may be a preferred option from an environmental and maintenance perspective, since tidal basins would reduce maintenance dredging and bring sediment to the beel areas. The tidal basin experience gained in the Khulna-Jessore Drainage Rehabilitation project may prove applicable in southwest Bangladesh, although the direct socioeconomic impacts need to be carefully considered. Pumped drainage seems a last and expensive resort, especially when the outside water levels become too high for gravity drainage. In any case, the proposed physical interventions will be extremely expensive. It is estimated that the cost of defending against a 1 m sea level rise would be on the order of US$1 billion (World Bank 2002). Lessons Learned from FCDI Development An evaluation of the effectiveness of flood control, drainage, and irrigation development was undertaken in the preparation of the National Water Management Plan, building on evaluations done for the Flood Action Plan and on other evaluations. A number of these have been highlighted with regard to the impact of embankments. Among the other lessons learned have been: · Implementation capacity has constrained project effectiveness. Delays in implementation have been frequent, often leading to cost overruns. Insufficient attention has been given to operation and maintenance of the ongoing and completed schemes, and some projects have been left incomplete (e.g. drainage networks). · Inadequate participation has affected every stage of the project cycle, from planning to implementation to operation and maintenance. · Lack of O&M has plagued virtually every public sector scheme evaluated. Its causes include insufficient government funding, poor cost recovery, lack of beneficiary participation, and technical difficulties. Based on figures from the Water Sector Investment Program (WSIP 1999), the National 23 According to the National Water Management Plan, capturing part of the sediment load of the major rivers to build up low- lying land could be one strategy to combat sea level rise, effectively using the natural delta-building processes. A rate of deposition that would match the rate of sea level rise may be feasible Functions and Technological Elements of Flood Control and Drainage Systems 47 Water Management Plan estimates (1994­98) annual Bangladesh Water Development Board maintenance expenditure on the order of Tk 1.2 billion, of which some Tk 1.1 billion would be for scheme maintenance. Required O&M expenditure is estimated at Tk 4.5 billion,24 more than three times actual spending. High siltation and erosion rates pose the most serious such problems. As a result of poor maintenance, long-term sustainability was judged doubtful. · Most projects initially achieved their technical objectives. Project design has often been satisfactory, except for drainage within flood control and drainage schemes. · Social impacts of past interventions have generally varied according to the incomes and livelihoods of the affected groups, but the environmental impacts of FCD have often been negative. Capture fisheries have suffered an especially negative impact. · Smaller projects have generally performed better than larger projects. Management of smaller units is less complex due to the smaller area and numbers of beneficiaries involved. · FCD and irrigation scheme net benefits are generally higher in the coastal regions than inland. This is because saline flooding is far more damaging than freshwater flooding, especially for agriculture, and because the scope for private irrigation in the coastal region is limited by its lack of shallow tubewell potential. Economic analysis of the FCD benefits undertaken in Flood Action Plan 12, concluded that the 17 inland projects showed an enormous range of economic internal rates of return (EIRR). Nine of the 17 projects were economically viable, with EIRRs of between 22 percent and 96 percent (median 54 percent), but 8 projects were not viable. Of these, 5 had EIRRs of 3 percent to 10 percent, one had a zero EIRR, and 2 were strongly negative (­10 percent). Smaller projects did somewhat better than larger projects (the median size of the 9 viable projects was 4,200 ha net cropped area) inclusion of an irrigation component improved economic returns, and pumped drainage appeared to be uneconomic (Ministry of Water Resources 2001). The Systems Rehabilitation project evaluation carried out in 1997­98 yielded comparable results. A cost-benefit analysis of 23 schemes indicated that 10 were uneconomic (EIRR lower than 12 percent). Most coastal polders, however, gave satisfactory EIRRs. All of this suggests that much of the production increase of the last 10 to 20 years can be attributed to the development of minor irrigation rather than FCD, through shallow tubewells and low-lift pumps. This is borne out by data collected during Flood Action Plan and NWMP studies: For a typical farmer, replacing nonirrigated crops with irrigated boro rice would more than double net income. Minor irrigation can thus do much to improve farm income(Ministry of Water Resources 2001). Options for FCD Improvement Options for improvement at FCD scheme level therefore seem to lie in: 24The cost of O&M is calculated as 3 percent of the scheme establishment cost, and three quarters of the O&M cost would be for maintenance, mostly earthworks. 48 Controlling or Living with Floods in Bangladesh 6. Improving and rehabilitating economically feasible schemes, focusing on their drainage function. This should be undertaken only where adequate maintenance provisions can be ensured, either through local-level management or increased government revenue allocation. 7. Ensuring adequate drainage provisions in all new development 8. Incorporating participatory planning and implementation approaches 9. Managing at the local level, incorporating the interests of all major stakeholders 10. Developing fish-friendly structures to improve connectivity between the FCD schemes, flood plain, and rivers 11. Developing flood proofing and preparedness measures and better adapted crop varieties in areas where rehabilitation or flood control are not desirable 12. Improving drainage capacity in drainage channels in coastal areas (e.g., through dredging, tidal basins) A number of these issues are further discussed in the next section. 5. Drainage and Institutions: A Policy Shift The availability of reliable data on the performance of FCD systems, an extensive learning process, and lively public debate (involving nongovernmental organizations, universities and public sector projects) have all contributed to a substantial shift in policy. The National Water Policy Under the National Water Policy, new FCDI development is no longer advocated but, in practice, it has not been sought since 1995. Emphasis is now placed on a "living with floods," based on flood proofing, improved flood warning and preparedness, and on development of flood-adapted varieties. Approved by the cabinet in January 1999, the National Water Policy resulted from a series of comprehensive consultations and debates, involving both the public and private (NGO) sectors. The main policy objectives are: · To address issues related to harnessing and developing all forms of surfacewater and groundwater and managing these resources efficiently and equitably. · To ensure the availability of water to every segment of society including the poor and the underprivileged, and to take into account the special needs of women and children. · To accelerate the development of sustainable public and private water delivery systems with appropriate legal and financial measures and incentives, including delineation of water rights and water pricing. · To bring about institutional changes that will help decentralize the management of water resources and enhance the role of women in water management. · To develop a legal and regulatory environment that will assist the decentralization process, ensure sound environmental management, and improve the investment climate for the private sector in water development and management. · To develop a state of knowledge and capability that will enable the country to design future water resource management plans by itself with economic efficiency, gender equity, social justice, and environmental awareness to facilitate achievement of the water management objectives through broad public participation (Ministry of Water Resources 2001b). The new policy marks a clear shift from infrastructure-oriented water management development toward improved use of existing resources with public participation and emphasis on management and knowledge rather than infrastructure development alone. In its policy, government defines its role as providing the regulatory framework while leaving a large role for the private sector to participate in implementing these guidelines. The government would undertake to: · Frame rules, procedures, and guidelines for combining water use and land use planning. · Frame and periodically revise the rules, procedures, and guidelines on all aspects of water management. 43 44 Controlling or Living with Floods in Bangladesh Moreover, the policy states that social and environmental assessments are mandatory in all plan development. The National Water Policy (Ministry of Water Resources 2001b) makes specific commitments to · Undertake comprehensive development and management of the main rivers through a system of barrages and other structural and nonstructural measures. · Develop water resources of the major rivers for multipurpose use, including irrigation, fisheries, navigation, forestry, and aquatic wildlife. · Desilt watercourses to maintain navigation channels and proper drainage. · Delineate water-stress areas, based on land characteristics and water availability from all sources to manage dry season demand. · Take steps to protect water quality and ensure efficient use. · Develop early warning and flood-proofing systems to manage natural disasters like flooding and drought. · Designate flood risk zones and take appropriate measures to provide desired levels of protection for life, property, vital infrastructure, agriculture, and wetlands. · Regions of economic importance such as metropolitan areas, seaports and airports, and export processing zones will be fully protected against floods as a matter of first priority. Other critical areas such as district and subdistrict towns, important commercial centers and places of historical importance will be gradually provided reasonable degree of protection against flood. · In the remaining rural areas, with the exception of those already covered by flood control infrastructure, the people will be motivated to develop different flood proofing measures such as raising platforms for homesteads, marketplaces, educational institutions, and community centers and to adjust the cropping pattern to fit the flood regime. · All national and regional highways, railroad tracks, and public buildings and facilities will, in the future, be constructed above the highest flood stage recorded in Bangladesh. This principle will also apply to reconstruction of structures of this type. · Make adequate provision for unimpeded drainage in all plans for roads and railroad embankments. · Undertake a survey and investigation of riverbank erosion and develop and implement master plans for river training and erosion control works to preserve scarce land and prevent landlessness and pauperization. · Plan and implement schemes to reclaim land from the sea and rivers. Drainage and Institutions: A Policy Shift 45 The National Water Management Plan The National Water Management Plan, finalized in 2001, is the first plan developed after the National Water Policy was issued and reflects the change in approach. The NWMP consists of main documents, supporting annexes, and a national and regional investment portfolio. The preliminary results of the NWMP, derived principally from the database of the Water Resources Planning Organization, expert consultations, and the Flood Action Plan studies, were initially discussed with a cross-section of the population in two rounds of public consultations in 24 locations in the first round and 28 locations in the second round. Subsequently, the main issues were worked out in a set of working papers that were again discussed with the main stakeholders in both the public and private sectors (Ministry of Water Resources 2001). Other Relevant Policies The National Water Management Plan, while based mostly on policy principles outlined in the national policy, also takes into account other relevant government-formulated policies. Policies pertinent to the drainage sector are: National Agricultural Policy (1999) Like former agricultural policies, the National Agricultural Policy of 1999 seeks to maintain self- sufficiency in food and to achieve the ultimate objective of poverty alleviation.25 Emphasis is placed on efficient irrigation and safe drinking water and, where possible from an environmental and social point of view, on increasing the development of groundwater irrigation. Specific attention is paid to promoting socially and environmentally friendly agriculture and maintaining a client-oriented agriculture system (National Agricultural Policy 1999 quoted on website Ministry of Agriculture, Ministry of Water Resources 2001). National Fisheries Policy (1998) The National Fisheries Policy of 1998 attempts to enhance fisheries production (especially where capture fishing is a main activity), boost exports, prevent further drainage of water bodies, and promote fisheries development in those bodies. On fisheries development, the Fisheries Policy and the National Wetlands and Wildlife Conservation Policy coincide. The stated policy to preserve existing water bodies and enhance fish production, limits virtually any new flood control, drainage, and irrigation development in Bangladesh, unless FCDI systems are equipped with clear mitigating measures like fish passes and other fish-friendly structures The main gap identified in the policy framework is for a national policy on land use or land use planning and zoning. The gap has been partly filled by Government Ordinance (June 21, 2001) concerning land use policy, prepared by the Ministry of Land. Although many operational aspects remain to be defined, the principle of land zoning is taken as starting point for development, thus providing a basis for further policy development. 25From a poverty alleviation point of view, one can debate whether the objective should be self-sufficiency in food production or sufficient purchasing power for the poor. 46 Controlling or Living with Floods in Bangladesh Public Sector Institutions Since water is such a crucial resource, some 35 government organizations, affiliated with 13 different ministries, are involved in the water sector. Although the public sector has played a prominent part in developing flood control and drainage, private sector organizations (e.g., equipment suppliers and contactors) and a vibrant NGO community play a growing role. Last but not least, direct stakeholders-- water users, farmers, fishermen, and local government officials--play a part. The Main Ministries The main government organizations involved in the water sector are · The Ministry of Water Resources, responsible for most aspects of the sector, including flood control, irrigation, drainage, water conservation, surface and groundwater use and river management. The Bangladesh Water Development Board falls under this ministry. · The Ministry of Local Government, Rural Development, and Cooperatives is organized into two divisions, the Local Government Division and the Rural Development and Cooperatives Division. Under the Local Government Division, the main implementing agency is the Local Government Engineering Department, which supports local government institutions at village (Union Parishad) and regional (upazila or thana) levels.26 There are 4,451 unions and 464 upazilas. Urban Drainage (sewerage and storm drainage), is the responsibility of the four major city corporations and the municipalities (Paurashava). · The Ministry of Agriculture, mainly through its Department of Agriculture Extension, deals with extension and, to a lesser extent, social organization of farmers and water users.27 · The Ministry of Fisheries and Livestock, through its Department of Fisheries, is responsible for capture and culture fisheries. · The Ministry of Land is mandated to develop and apply land-zoning principles and is the custodian of all government-owned land and water bodies. · The National Water Resources Council is the highest national body reporting to the cabinet on all water issues. Its executive committee provides direct guidance and directives to the operational departments and organizations in the water sector. · The Water Resources Planning Organization is responsible for water sector planning, drafting and updating the National Water Management Plan, and maintaining the water resources database. WARPO acts as secretariat to the executive committee of the National Water Resources Council. · The Planning Commission is a national body responsible for screening and approving all development projects, drawing up the Five-Year Plan and the Annual Development Program. 26Zila, District Parishads, are new bodies that are to be formed at district level, elected from Upazila Parishads and municipalities. These have yet to be elected. 27Agricultural research is carried out under the umbrella of the Bangladesh Agricultural Research Council by its 10 affiliated research institutes. Drainage and Institutions: A Policy Shift 47 · The Ministry of Environment and Forests, with its Department of Environment, is mandated to regulate and enforce environmental management, including pollution control, and ensure the adequacy of Environmental Impact Assessments. Though extensively mandated to be involved in the water sector, the Department of Environment has so far been severely constrained by a lack of an operational budget. The main government organizations involved directly in drainage development are discussed below. The Bangladesh Water Development Board The Bangladesh Water Development Board is responsible for planning, construction (including rehabilitation), and operation and maintenance of larger structures in FCD systems like embankments, sluice gates, regulators, bridges, and roads. The board, which was known as the Water and Power Development Authority (WAPDA) before independence, was split up in 1972, leading to the constitution of the Bangladesh Water Development Board and the Bangladesh Power Development Board.28 Since then, the BWDB has been responsible for the overall development and management of flood control, drainage and irrigation projects in Bangladesh. Between 1994 and 2000, by implementing some 70 projects, the BWDB was responsible for 85 percent of all investment in the water sector, with the Ministry of Agriculture, the second biggest investor, responsible for an estimated 8 percent. Major donors have been the International Development Association, the Asian Development Bank, the Netherlands, and Japan. An overview is provided in table 18. Table 18 Expenditures in the water sector Organizations Non-BDWB Total BWDB MoA LGED BMDA BADC Total project value x 1,000,000 Tk 72,649 7,871 6,131 5,516 6,301 98,468 Percent 74 8 6 6 6 100 Disbursement (by financial year) 1994­1995 6,097 1,497 7,594 1995­1996 6,749 1,111 7,860 1996­1997 7,959 904 8,863 1997­1998 8,250 754 9,004 1998­1999 6,456 694 7,150 1999­2000 4,364 2,002 6,366 Total 39,875 6,962 46,837 Percent 85 15 100 BWDB Bangladesh Water Development Board; MoA Ministry of Agriculture; BADC Bangladesh Agricultural Development Corporation; LGED Local Government Engineering Department; BMDA Barind Multipurpose Development Authority. Source: Ministry of Water Resources 2001. 28The BWDB is still referred to as pani WAPDA by many farmers (pani = water). 48 Controlling or Living with Floods in Bangladesh Until the Local Government Engineering Department became involved in flood control and drainage in the beginning of the 1990s, the BWDB was the only government institution dealing with water resource development. The BWDB has developed some five hundred FCD schemes over the last 30 to 40 years, covering an area of about 6 million ha (table 18). As discussed in this report, the effectiveness of the flood control, drainage, and irrigations schemes developed has been subject to considerable debate, culminating in a major shift in national policy. To be able to implement the changes identified and increase the effectiveness of the organization, a reform process was initiated,29 in line with the recent policy and development plans and guidelines. The Reform Process The Water Development Board has been cutting staff in recent years, halving its payroll from 18,000 employees in 1990 to fewer than 9,000 in 2002. The reform process is formalized in the Bangladesh Water Development Board Act of 2000, requiring that board functions be guided by the National Water Policy and the NWMP. The act confirms the principles as set out in the National Water Policy and the Guidelines for Participatory Water Management for the transfer of BWDB projects up to communities, local government institutions, or both. Other elements of the reform process include: · Downsizing to approximately 8,000 employees and reorientiing staff composition to include close to 1,000 nonengineering professionals · Incorporating participatory planning procedures · Rationalizing flood control, drainage, and irrigation schemes and transferring management of viable schemes to local governments and beneficiary organizations · Instituting a board of directors, drawn from both the public 30 and the private sector, to whom the director general (formerly the chairman) would report · Appointing the director general for a minimum period of three years. Three programs have been specifically designed to assist the BWDB in the reform process; the Water Sector Investment project, focusing mainly on inland areas; the Integrated Planning for Sustainable Water Management project,31 focusing mainly on coastal areas and on participatory planning and incorporating the principles of the Guidelines for Participatory Water Management and a recently started "twinning program," running from 2001 to 2006, with the Netherlands Ministry of Transport and Water. The Water Sector Investment project, under consideration for financing by the World Bank, would provide support to the BWDB to enact the new guidelines for management transfer and assist the BWDB in becoming a financially autonomous organization, as stipulated in the National Water Policy. There is no agreement yet on investment projects, and its implementation is not assured. 29Much of the impetus for reform was initiated earlier, through programs such as Flood Action Program and the Systems Rehabilitation project. These originated as highly technical and construction-oriented projects and later emerged as driving forces behind institutional reform 30Including other Ministries such as the Agriculture, Fisheries, Environment, and Planning 31For IPSWAM, see Government of the Netherlands, 2002 Drainage and Institutions: A Policy Shift 49 Virtually no new schemes have been developed since 1995. The focus of the BWDB has been mostly on rehabilitation except for minor developments in the Northeast (Dampara Water Management project with funding from the Canadian International Development Agency) and in the newly settled areas of the Southeast (Char Development and Settlement project, financed by the Netherlands). Ongoing projects include the Asian Development Bank­financed Jamuna-Meghna River Erosion Mitigation project, the Command Area Development project, the Pabna Irrigation and Rural Development project, the Meghna- Donagoda Irrigation project, Khulna Jessore Drainage Rehabilitation Program , the Integrated Coastal Zone Management Program financed by the Netherlands' government, and the Coastal Embankment Rehabilitation project. The Water Resources Planning Organization The Water Resources Planning Organization was an outgrowth of the Master Planning Organization in the 1980s. The Master Planning Organization prepared the first Water Resources Master Plan, in 1986, and played a role in most of the subsequent plans. WARPO is mandated to prepare, and periodically update, the National Water Management Plan. Although the cabinet has not yet formally approved the plan, finalized in 2001, it is being used as a guideline. WARPO is mandated to screen all proposed water-related projects in the light of the NWMP, the National Water Policy and relevant guidelines and, as mentioned, acts as secretariat to the executive committee of the National Water Resources Council. WARPO received extensive support during the previous NWMP formulation process, which has enabled it to assume a more central position than in the past. The National Water Policy stipulates that WARPO should retain its central role as clearing house and main decisionmaker for water resources planning and development, also after the NWMP project has been finalized. WARPO is the author of the recently drafted water code, the absence of which was identified as a major bottleneck to further institutional development in the drainage sector in both the water policy and the master plan. The Local Government Engineering Department The Local Government Engineering Department originated in the early 1960s when implementation of a Rural Works Program commenced. In 1982, a Works Program Wing was established in the Local Government Division of the Ministry of Local Government, Rural Development, and Cooperatives and converted to the Local Government Engineering Bureau in 1984. Increased budgets and the expansion of the scope of works prompted upgrading to the present department level in 1992. The department has attracted substantial donor funding in the last decade and is currently responsible for 36 development projects involving a total cost of US$1.2 billion. Its position as a government organization at the local level, its sensitivity to donor and community concerns, and its emphasis on community-based, participatory development made it an attractive alternative to many donors. The department has grown into one of the largest public organizations in the rural sector. Its employees number some 9,600, 89 at headquarters, 726 at district level, and 8,785 at upazila level (2001). Involvement in FCD The Local Government Engineering Department became active in the development of flood control and drainage in the mid-1980s, first on a minor scale, then gradually increasing to its current level. The Small Scale Water Resources Development Sector Program has been the department's most extensive program in water resource development. The program was initiated in 1995 with financing from the Asian 50 Controlling or Living with Floods in Bangladesh Development Bank, the International Fund for Agricultural Development, and the Netherlands government. The program is concentrated in the western half of Bangladesh, and participatory planning and implementation principles were included in its design and methodology. Subprojects are identified and proposed by beneficiaries through local government and finally selected by the Local Government Engineering Department. Beneficiary organizations, called water management cooperative associations, are developed to help plan, implement, operate, and maintain the schemes. Each association is required to contribute 3 percent of earthwork capital costs, 1.5 percent of structure costs, and 30 percent of land costs, and to bear the full O&M costs. At least two thirds of the subprojects are based on existing flood control and drainage infrastructure developed by the Bangladesh Water Development Board (Ministry of Water Resources 2001). The main elements of the small-scale water development approach are: · Involvement of local government organizations through project implementation committees in identifying, setting priorities, and planning interventions · Involvement of Labor Contracting Societies, fashioned on the Landless Contracting Societies developed under Early Implementation projects in the 1980s and 1990s, in the execution of the work, targeted at the poor · Involvement of NGOs in community mobilization · Emphasis on small schemes, partly by dividing larger, BWDB­developed schemes into smaller, more manageable units · Development of a clear, step-by-step approach with a defined role for beneficiary participation, in decision-making and financial contribution (cash or kind). Despite its innovative and participatory design, an interim evaluation, commissioned by the Dutch government and the Asian Development Bank, showed that the issues faced in the BWDB­implemented schemes were also present in engineering department schemes. Points of concern included: the focus on participation for implementation rather than for management (O&M); and mixed construction and design quality, which would hamper smooth O&M by the community organizations, even in the short run. These points are reminiscent of the reasons given for the underperformance of larger flood control and drainage schemes (Chowdhurry and van Immerzeel 1999). An in-depth evaluation, scheduled for the end of 2002, was expected to provide further lessons for the future program of management transfer and FCD development at a national level. Agricultural Extension The Department of Agricultural Extension, through its extensive cadre of block supervisors, is the main local arm of the Ministry of Agriculture. With support from U.K. Department for International Development and the World Bank, the Department of Agriculture formulated an updated New Agricultural Extension Policy and set up a Water Management Wing, staffed by many of employees made redundant after dismantlement of the Bangladesh Agricultural Development Corporation. The department focuses on on-farm water management and irrigation agronomy at field level and implements policy through coordination committees at regional,32 district, and thana level. The regional committees also 32The Department of Agricultural Extension has divided Bangladesh into 18 regions. Drainage and Institutions: A Policy Shift 51 include representatives from NGOs and the research establishment. A cadre of 12,600 block supervisors, each maintaining links with 900 to 1,200 farmers, implements its policies at field level. Although the block supervisors are the only government staff with a specific mandate in agriculture (and water management), their minimal operational budget and decision-making powers limit their effectiveness at the farmer level. Local Government Institutions: Union Parishad and Upazila Parishad The Union Parishad is the lowest level formally elected local government institution. From 7 to 12 villages fall under its jurisdiction, and each union covers from 10,000 to 20,000 ha. Thus, small and medium FCDI schemes usually fall under the jurisdiction of one or two unions. The Union Parishad is officially responsible for planning implementation of social and physical infrastructure development activities in the area. Attempts to strengthen the union have had limited success, due partly to the highly political nature of such support programs. Nevertheless, LGIs such as the unions are the only formally elected (thus democratically accountable) organizations in the rural areas. A major constraint for unions is their lack of authority and resources to undertake development activities, for example, in FCDI schemes. Despite these limitations, unions still play a role in infrastructure development, using resources allocated by the central government. They are not yet directly involved in water management, but they facilitate conflict resolution and resource mobilization and influence decision-making. Moreover, LGIs are directly involved in Food for Work Programs. Nongovernmental Organizations Regional, national, and international NGOs have been involved in the water sector, both as partner and as vocal critic of flood control and drainage developments. Before and during the Flood Action Plan, NGOs were not usually involved directly in FCD development programs and focused more on their traditional strengths: credit provision for the poor, income-generating micro projects, relief work, development of social facilities, as well as advocacy and human rights. NGOs such as Grameen Bank and BRAC have attained international recognition, and concepts developed in Bangladesh such as group credit have been copied across the globe. It is generally accepted that NGOs have made a substantial contribution to improved living standards in Bangladesh and have helped the rural population cope with the disasters caused by floods and cyclones. In the process of micro credit operation, NGOs have organized groups, village centers, and village organizations. Such groups are formed by marginal and poor male and female members and, using micro credit, many group members have improved their livelihoods and invested in small businesses, livestock, and agricultural activities. In emergencies, these local groups mobilize resources for repairs and maintenance. Attempts to involve NGOs in flood control and drainage work have had mixed results. NGOs have been directly involved in community mobilization and participatory planning in projects such as the Systems Rehabilitation project, the Early Implementation project, and the Small Scale Water Resources Development Program. 52 Controlling or Living with Floods in Bangladesh The Cooperative Movement Since the 1960s, 63,000 farmer societies, Krishi Shamabya Samity (KSS),33 have been set up in Bangladesh, with mixed results. Though formally a private sector organization, KSS have survived with the help of government subsidies and programs but have had a limited role as genuine cooperative organizations. Apart from KSS, other cooperatives were also active in the 1960s and 1970s, mostly in the areas of credit and savings, but NGOs have taken over their role. Private Irrigation Suppliers Once the exclusive province of the Bangladesh Agricultural Development Corporation, the small-scale irrigation sector was deregulated in the late 1980s and early 1990s. A vibrant private sector has developed, and private companies, pump and engine suppliers, and drilling companies supply the necessary services. More than 500,000 tubewells and 10 million hand pumps for drinking water have been installed through the private sector (Ministry of Water Resources 2001). The current market is estimated at US$30 million annually. Project-User Committees for Water Management A multitude of project-related user committees and ad-hoc organizations have been set up in the past two decades. Often, these committees were initiated by the projects to facilitate planning and implementation and, later, to undertake minor maintenance and take part in construction. A number of projects have involved local contracting societies, especially in the earthwork, as income-generating activity for the poor. The National Water Policy now requires that 25 percent of the earthwork on any public water project should be offered to specific target groups or beneficiaries, for example, the landless or women- headed households. Water Management Group, Water Management Association, and Water Management Federation The Bangladesh government has recently adopted a policy of making water user groups responsible for water management and O&M. A three-tier organization is foreseen, consisting of water management groups, water management associations, and water management federations. The principles governing the establishment of these organizations are laid down in the National Water Policy and worked out in the Guidelines for Participatory Water Management. The water management group is the lowest tier of the user management organization, typically covering one village or a small hydrological unit. Membership is open to all farmers, traders (small or big), craftsmen, boatmen, fishermen, the landless, and destitute women within the subsystem or subproject. Members have to pay an admission fee, purchase at least one share, and contribute to the costs of water management. The water user group is responsible for collecting any fees and handing them to the next tier, the water management association. User group representatives make up the water management associations' general assemblies whose representatives form the general assemblyof the water management federation, the highest tier. The guidelines foresee that, in small schemes (less than 1,000 ha), a number of water management groups and one water management association would be responsible 33The KSS evolved in the mid-1960s as a rural development approach of the Comilla Rural Development Academy. The academy tested and implemented innovative rural development approaches in the 1960s. Drainage and Institutions: A Policy Shift 53 for management. In schemes larger than 5,000 ha, a water management federation would be established. For schemes between 1,000 ha and 5,000 ha, a federation could be established according to need. The guidelines provide for the registration of each water management association as a cooperative, with its own bylaws, in accordance with the Cooperative Societies Ordinance (1984) and its subsequent amendment and the Cooperative Societies Rules (1987). The concept of water management cooperative associations has been in place since 1987 when the Local Government Engineering Department pioneered this form of organization. From 1995­96 on, all projects funded by the Asian Development Bank and executed by the Bangladesh Water Development Board and the Local Government Engineering Department established water management cooperative associations in their project areas. The Dutch-funded Systems Rehabilitation project (closed in 1998) played a leading role in further developing the concept of water management organizations and provided the main impetus for the development and acceptance of the Guidelines for Participatory Water Management. Sluice Committee­Inlet Committee Sluice­inlet committees were among the first local institutions organized to operate and maintain water control structures for flood control, drainage, and irrigation projects. Sluice committees, consisting of local leaders, were usually formed with the support of the executing agencies. Such committees were organized both by the BWDB and the engineering department. The Local Government Engineering Department discontinued this approach after developing the water management cooperative concept in 1987. The BWDB continued with sluice committee formation until 1998. Most sluice committees were not active after the initial formation period, and few exist today. Wetland Management Organizations During the last decade a number of projects have fostered the development of user organizations. The Management of Aquatic Ecosystem through the Community Husbandry project and the Community Based Fisheries Management project have been particularly active in developing user organizations for the management of seasonal or perennial wetlands (haor/baor/beel). These organizations, called resource management organizations or beel management committees are responsible for the maintenance of sanctuaries and reexcavation of canals and depressions and are active in Sherpur, Gazipur, Molvibazar, and other districts. These projects are being implemented by the Department of Fisheries with substantial NGO involvement. Informal Local Water Management Arrangements A number of cases have been documented of local arrangements for operation and maintenance of irrigation and drainage infrastructure without any formal organizations. Duyne describes how intricate cooperative arrangements are put in place, involving considerable quantities of labor and cash and a high level of internal organization (Duyne 1998). Though much of these traditional arrangements occur in the Northeast, in the haor area, similar arrangements can be found in other parts of Bangladesh. There are some indications that local management traditions were more developed in the past. In one of the scheme areas of the Early Implementation projects, Blok (personal communication) found that, until the 1970s in Dirai/Sunamganj, a local organization managed the cross-bundhs within an FCD system. This organization had local rules, a rudimentary court, and mechanisms to ensure a basic equity. The organization reputedly eroded as a result of the large migration movements of the 1970s and the arrival of 54 Controlling or Living with Floods in Bangladesh political party concerns on the scene. Another case was found in the Northwest, in Naogaon Polder, as described in box 2. Box 2 Compartmentalization in Polder C An interesting example of a local flood control and drainage initiative was found in a 3,000-ha farmer- operated mini-polder in one corner of Polder C. By heightening an existing road embankment, the local community made its mini-polder independent of the main polder. Four local committees were formed, each looking after its own beel and its own stretch of embankment. These committees work closely together when necessary. The one drainage regulator is operated for drainage or irrigation storage, as desired. Maintenance is funded by a Tk150 per ha farmers' contribution, a five-year fishing lease for Tk 200,000, and emergency contributions by big landowners when required. The mini-polder escaped damage in the 1998 flood. Essentially, the mini-polder farmers have adopted a "living with the flood" approach, with a form of controlled flooding. Rather than attempting to replace the original B. (deepwater) aman with LT or high-yielding T. aman, they have minimized flood depths, which are maintained at levels not much below those outside. This enhances embankment safety. B aman remains the dominant monsoon crop (75 percent now, as compared with 85 percent preproject), but fish production is maintained. Irrigation boro is now the priority crop, as in the rest of Polder C. Source: National Water Management Plan 2000. Financial and Legal Aspects Financing In the drainage sector, budget shortfalls for maintenance have been a major threat to sustainability. The greater part of the Bangladesh Water Development Board budget is allocated for land tax (paid to the Ministry of Land), and Bangladesh remains dependent upon external aid34 to finance a large part of its expenditures. Based on data from the Water Sector Investment Program, the National Water Master Plan estimates for1994­98 foresaw annual maintenance expenditures of Tk 1.21 billion, of which some Tk 1.1 billion would be for maintenance. Required O&M expenditure is estimated on the order of Tk 4.5 billion,35 more than three times what was actually spent. Except for the irrigation systems in the Barind Agricultural Development project, cost-recovery rates for irrigation services are low. Moreover, the applied rates are far below actual expenditures (Ministry of Water Resources 2001). According to the National Water Policy, cost recovery would be sought only for public irrigation services, not for flood control and drainage services.36 Poverty, a dependence on external aid, and the free-rider problem are used to explain the presumed lack of willingness to pay for services. Although users do not formally contribute to capital investments they do, de facto, execute and finance the major part of operation and maintenance activities, not only in small-scale irrigation but also in larger drainage and flood control works. The amount users spend on O&M is not formally recorded but has been 34In 2000, external economic aid amounted to US$1.6 billion. State revenues amounted to US$4.9 billion for FY1999/2000 . 35Whereby the cost for O&M is calculated as 3 percent of the scheme establishment cost; three quarters of the O&M cost would be for maintenance--mostly earthworks. 36"For the foreseeable future, . . . cost recovery for flood control and drainage (FCD) projects is not envisaged in this policy" (National Water Policy, Sec. 4.14,: 15­16, Ministry of Water Resources 2001). Drainage and Institutions: A Policy Shift 55 documented in a number of studies. Duyne (1998) recorded more than 700 discrete activities in her sample of four flood control, drainage, and irrigation systems (representing coastal, haor, and beel systems) throughout the country. Users contribute both in kind and cash, up to Tk 400,000 for community infrastructure (in this case for embankment construction in a haor area in the Northeast) and Tk 1,500 to 4,000 for private, individual activities such as installing drainage outlets. Local government (Union Parishad) or community funds, raised through lotteries or fairs, are also utilized for maintenance and repair. Informal organizations have proven highly effective in mobilizing local resources (labor, materials, and cash) and in quickly undertaking necessary repairs (Duyne 1998). Box 3 illustrates user willingness to pay. Box 3 Water charges collected by farmers An illustrative case of ability and willingness to pay is the Karnafouli Irrigation Project in the Chittagong area. The project was rehabilitated in the mid-1990s, when the Guidelines for Participatory Water Management were still being developed. As part of the rehabilitation process, farmers were organized into water management groups and water management associations. The farmer representatives agreed to pay irrigation charges according to government rules with the guarantee that the funds would be retained locally. A joint Bangladesh Water Development Board­water management association bank account was opened, and in the first few years nearly Tk 2 million was collected from the farmers and deposited into the account. However, a clear, legally accepted mechanism for setting priorities, disbursement, and control had not been fully developed when the project ended, and the money remains unspent to this day. No wonder farmers in the Karnafouli project aren ow reluctant to pay irrigation charges. Source: Own research. These cases show that: "People have both the capacity and the willingness to contribute substantial amounts of money, labor, and materials for development, maintenance, and operation (...), provided that they are able to recognize a direct link between their contribution and the service received" (Duyne 1998: 51).37 The case presented in box 4 illustrates this point. 37The authors have similar experience in Pakistan and Egypt where farmers are willing to pay for services they need and can control. They are not willing to pay --and maintain that they are not able to pay--for services they cannot control. 56 Controlling or Living with Floods in Bangladesh Box 4 Tendering for irrigation in Naogaon The flood control and drainage project in Naogaon Polder-1 was developed by the Bangladesh Water Development Board with assistance from the International Development Association. In Polder 1, there are two irrigation arrangements: groundwater irrigation by deep tubewells, operated by Barind Multipurpose Development Authority in the high areas, and surfacewater irrigation from the adjacent river by privately operated low-lift pumps in the lower areas. The river is not navigable in the dry season. During this time, the riverside villages call a meeting in a large open place, and anywhere from 5,000 to 7,000 thousand people gather. In this meeting, led by the village elders, decisions are made on the number and location of the cross-bundhs and water rotation between them. A public auction is also held for water supply services. Pump owners bid for supply contracts, and the lowest bidder wins the "contract." Afterward, people go to their designated section of the river to construct their own bundh with their own labor. Once the main bundh is constructed at the lower end, villagers take water by low-lift pumps for two weeks. Then a second bundh is constructed, and people at the upper end take water the next two weeks for transplantation of paddy. This is the traditional way lower and upper end villagers share water from the river. The water supplier is responsible for supplying enough to grow rice and maintain the distribution canals. To ensure sufficient water supply, they keep one pump set as a standby. Villagers assist the water supplier in collecting the water charges on time, and sometimes villagers negotiate with defaulters for payment of water charges to the water supplier. The water supplier estimates the water charge in relation to distance from the water supply, the condition of the irrigation canals, and the need to transport his equipment. Source: Own research. Legal Framework A series of acts, circulars, rules, laws, guidelines and ordinances form the present legal framework for the drainage sector.38 The main acts with respect to flood control, drainage, and irrigation systems are: · Bengal Cooperative Society Act (1940), providing for the organization of cooperative societies that are especially concerned with irrigation, drainage and flood protection · State Acquisition and Tenancy Act (1950), regulating the appropriation of privately owned land, for example, for embankments and roads · Embankment and Drainage Act (1952) and its subsequent amendments. The act vests in the government of Bangladesh or its respective authority (the Water Resources Planning Organization and later the Bangladesh Water Development Board) the power to take over, for example, public embankments and watercourses, and provides rules for the construction of temporary dams, roads, and watercourses and opening and shutting of sluices. · Water Board Act (2000), regulating the board's structure, funding, and policymaking authority. · Other legal documents relevant to the sector include: · Cooperative Societies Ordinance (1984) 38Existing legal provisions for water management will be superseded by a new water code being discussed at policy level. The code will incorporate the main new elements of the National Water Policy. Drainage and Institutions: A Policy Shift 57 · Cooperative Societies Rules, legalizing the establishment of the Water Management Cooperative Association (1987) · National Land Use Policy Ordinance (2001) · Irrigation Water Rate Ordinance (1983) · Local Government (Union Parishads) Ordinance (1983) · National Policies for Water, Agriculture, Fisheries, and Environment · Guidelines for Participatory Water Management, Project Formulation, and Environmental Impact Assessment FCD Management Arrangements in the National Water Policy and Guidelines The transfer of management over schemes smaller than 5,000 ha is foreseen in the National Water Policy (and Water Board Act 2000) from the Bangladesh Water Development Board to local government institutions and community groups. According to the Guidelines for Participatory Water Management, these communities, organized into a two-to-three tier organization, would manage their scheme in cooperation with local governments (table 19). Schemes between 5,000 ha and 15,000 ha would be managed jointly by the BWDB and the LGIs/communities, with possible involvement of the private sector. Management of schemes larger than 15,000 ha would remain the BWDB's responsibility. Details have not been worked out on the organization of cooperation and joint management or the legal ramifications and would be the focus of projects such as the Water Sector Investment Program (mostly inland areas) and the Integrated Planning for Sustainable Water Management (mostly coastal areas). Table 19 Institutional models in the Guidelines for Participatory Water Management Size Ownership Management arrangement < 1,000 ha Local government institution (LGI) LGI 1,000 ha to 5,000 ha LGI/Bangladesh Water Development Board LGI/community (BWDB) > 5,000 ha BWDB Private sector/ LGI/ community/BWDB Source: Ministry of Water Resources 1999. Comparison of Organization Models The guidelines provide a major step forward in acknowledging the need for stakeholder participation in FCD development and providing a framework for institutional reorganization. In the National Water Master Plan, an additional organization model is proposed, the nonprofit management company, which could be similar in its set-up to a Water Management Board as known in the Netherlands. Such an organization could be financed through a land improvement tax, depending on the level of drainage, flood protection, and water availability for irrigation. An appropriate dividing line between the different levels of management arrangements, instead of size, might be water management functions, interests and roles, and implementation capacity. The Bangladesh Water Development Board could, for example, remain responsible for major interventions and 58 Controlling or Living with Floods in Bangladesh infrastructure such as seaside and major riverside embankments, major regulators, and dredging. Local organizations could assume responsibility for minor infrastructure such as embankments, structures, and internal drainage networks. Piloting and a phased approach would allow local organizations to build up capacity, and progressively larger responsibilities could be transferred to them, accompanied by targeted subsidies, instead of providing a blanket subsidy for O&M. The main options are set out in table 20. Table 20 Institutional models for managing flood control, drainage, and irrigation Model Legal model Funding O&M Notes Participation Bangladesh Water User groups BWDB Users generally unwilling by water user Development Board contribute in kind or remains to contribute without groups (BWDB) retains cash on a voluntary responsible some form of control. formal authority and basis, e.g. through and solicits Success requires pays the annual land matching user substantial involvement in development tax to contributions participation project or with the Ministry of Land nongovernmental organization (NGO) Multitier Cooperative obtains Member Cooperative No guarantee that cooperative usufruct rights from contributions, bank cooperatives will government while loans, embankment represent all farmers and tenure remains with leasing, subsidies interests (fisheries). BWDB or is Membership is voluntary: transferred to local enforcement of rules on government nonmembers has no legal basis. Local Complete hand-over Introduction of new Local LGIs generally lack government to local government land improvement government, in resources and skills. institution (LGI) that tax. cooperation Accountability relations pays annual land Embankment leasing, with user to users doubtful development tax subsidies groups Nonprofit Local government A new land Users, Similar to water board management would form a improvement tax, professional model. Need to build up company or nonprofit company based on land classes staff, completely new (local) water under new legislation or potential, and use contractors institution representing management with elected company of water bodies. potentially conflicting board directors or board Embankment leasing, interests. members from the targeted subsidies area Source: Ministry of Water Resources 2001; own research. 6. Conclusions This case study on agricultural drainage in Bangladesh was done as part of the project: Agricultural Drainage: Toward an Interdisciplinary and Integrated Approach. The study, carried out in summer 2002, is based on both desk research outside Bangladesh and a three-week field study in the country. Bangladesh was selected as the case of agricultural drainage in humid tropical environments. The main conclusions of the study follow. The Need for Agricultural Drainage The type of drainage that has developed in Bangladesh is largely conditioned by its humid climate and position at the mouth of three major rivers, of which two are transboundary. Bangladesh is a large, flat and highly active delta under considerable tidal influence, where 30 percent of the land is flooded each year. For agriculture, the primary function of flood control and drainage schemes (embankments and drainage regulators) is to reduce the flooding depth of protected land to enable a higher crop yield, mainly rice. Additional functions include the protection of boro rice crop from early flash floods in the inland areas and the exclusion of saline water in the coastal areas. There are three types of drainage: · Gravity drainage under tidal conditions, practiced in the coastal areas where gravity drainage is possible only at low tide · Gravity drainage under nontidal conditions, practiced in the inland areas and at the end of the monsoon season when waters outside the embankments recede · Pumped drainage, on a limited scale, generally developed with the dual purpose of irrigation and drainage. Typical Characteristics: Multiple Interests, Functions, and Stakeholde rs Agricultural drainage in Bangladesh should be understood as water control in the broadest sense of the word. Four characteristics stand out: · Conflict among the different groups of land and water users · Complex management of water levels and quality in areas with competing and mutually exclusive demands on water levels and, in the coastal areas, salinity · Diversity of drainage conditions, even regionally · Variety of functions of flood control and drainage infrastructure such as protection of homesteads, agricultural land, and drainage (their primary functions), irrigation, navigation, transport crossing embankments, flood shelter, and housing 59 60 Controlling or Living with Floods in Bangladesh Five main FCD types or land forms correspond to different geomorphological conditions and manmade interventions. Analysis of the land forms and the hydrological region leads to the definition of the main drainage issues in Bangladesh (table 21). Table 21 Summary of main drainage issues in Bangladesh Land form- Hydrological region Main drainage issues FCD System Deltaic coastal Southwest, Central Drainage under tidal conditions polder and East Control of water levels inside polder for different needs Control of salinity; balance interests shrimp and rice Maintenance of drainage capacity outside polders; siltation reduces drainage base. Nondeltaic Eastern Hills Control of salinity, balance interests agriculture­salt coastal polder production­shrimp growing. Management of drainage under tidal influence with cyclones and tidal surge Beel Northwest, North Monsoon flooding by rainfall and runoff Central Balance beel preservation for fisheries with drainage for agriculture. Drainage constrained due to high river water levels and silting up khals Flood plain Across the country Congestion due to insufficient drainage outlets in except Eastern Hills embankments and roads and Northeast Confinement caused by embankment, increased flooding in unprotected areas Hoar Northeast Timing of embankment cutting to balance interests of fishermen, boatmen and agriculture Source: Own research. FCD Has Impaired Adequate Drainage From the 1960s on, a massive nationwide campaign for flood protection and agricultural development took place through the development of flood control and drainage schemes. Some 6 million ha have been covered by FCD schemes, and most of the flood-prone land where FCD schemes would be viable has been developed. The development of embankments and other infrastructure has impeded the natural drainage patterns, in some cases contributing to, rather than alleviating, local drainage congestion--in both coastal and inland areas, and siltation of drainage channels regionally, in the coastal areas. FCD Scheme Performance Has Been Mixed The most positive impact of FCD development has been in the coastal regions where there are fewer opportunities for groundwater irrigation and the negative effect of saltwater flooding is higher than in inland systems. In other FCD schemes, less positive results were obtained. Studies indicate that economic returns on about half the FCD schemes surveyed were satisfactory (BWBB 1998). Small schemes (median size of the most successful schemes, 4,200 ha) performed better than large ones. The most Conclusions 61 negative impacts have been social and environmental in nature, with increased conflicts over water management and reduction of capture fisheries. External FCD impacts--decreased capture fisheries and raised external flood levels--seem to have cancelled out the positive impacts on a regional level. Policy Has Shifted from Flood Control to Living with Floods As a result, a marked change in approach has taken place. The new National Water Policy (1999) emphasizes living with, rather than controlling, floods and stresses the incorporation of social and environmental concerns in all FCD development. Major emphasis is placed on flood proofing, minimal disruptions to the drainage network and fish migration patterns, and disaster preparedness. The National Water Policy and subsequent National Water Management Plan identify the following priorities for future development. · Improving and rehabilitating economically feasible flood control and drainage schemes but only where adequate maintenance can be ensured · Ensuring adequate drainage provisions in all new development · Incorporating participatory planning and implementation approaches · Providing for local level management, incorporating the interests of all major stakeholders · Developing fish-friendly structures to improve connectivity between the FCD schemes, flood plain, and rivers · Developing flood proofing and preparedness measures and better adapted crop varieties in areas where rehabilitation or flood control is not feasible or desirable · Improving drainage capacity in drainage channels in coastal areas (e.g. through dredging, tidal basins). Institutions Financing maintenance is possibly the biggest bottleneck to sustainability. Only a third of what is needed for proper maintenance is now being spent. Most FCD schemes are caught in a maintenance trap, requiring regular rehabilitation to perform their functions. Users have made substantial investments in FCD development when they have been able to exercise control over designation of priorities and use of funds. Participation A lack of a participatory approach has been identified as a major constraint to sustainable FCD development. Small schemes developed with close involvement of the beneficiaries seem to be more successful and may offer a viable alternative to future FCD development. The government's Guidelines for Participatory Water Management foresee extensive stakeholder involvement in project planning and implementation and transfer of management to community organizations and local government institutions. Experiments with these forms of local-level management have recently been taken up, initially in irrigation projects and subsequently in flood control and drainage schemes as well. Recognizing the need for close linkage between financing and decision-making, and the need to weigh multiple diverse interests in water management, the water board model may be appropriate in Bangladesh. Glossary Aman. Monsoon season paddy (rice) crop Aus. Late dry season/early monsoon paddy (rice) crop Beel. Naturally occurring (semi-)permanent water bodies formed in a low-lying depression in the flood plain that fill up and expand in the monsoon season, becoming huge lakes, and decrease or dry up in the dry season Boro. Winter (dry) season paddy (rice) crop Bundh. A small earthen embankment or dam Cross-dam. A body of earth placed across a khal or river for retaining water Drainage regulator. A regulator placed in an embankment with flap gates on the river or sea side Embankment. A wall or ridge of earth to protect an area from flooding or to carry a road or highway over a low-lying area Embankment, public cuts. The practice of breaching an embankment to allow congested river and rain water to be drained from an area, typically toward the end of the monsoon season when river levels start to drop Fall boards. Boards placed in slots or grooves in the pier walls of regulators or sluices to close the vents for maintenance or water retention FCD system. Flood control drainage system. All the areas in the flood plains of the rivers and in the coastal plains utilized by humans and containing some or all of the following infrastructure: khals, beels, cross-dams, canals, embankments and regulators FCDI system. Like FCD, but with measures for the supply of irrigation water, including khal excavation and with or without hydraulic structures Flap gate. A swinging gate on the river- or seaside of a regulator or sluice vent that automatically closes when the water rises above the country-side water level Flushing. The practice of letting in (fresh or salt) water for irrigation (or shrimp production) through regulators or inlets Haor. An extensive depression between the natural levees of rivers, saucer-shaped, with a deep central part permanently under water, found in Northeastern Bangladesh Katcha. cross-bunds Khal. A natural channel, minor river, or a tidal creek 63 64 Controlling or Living with Floods in Bangladesh Kharif. Cropping period during the wet season (May through October), divided into kharif-I and kharif-II Khas. Land owned by the state, including recently accreted land Monsoon. The rainy season, starting in June and ending in October Natural levees. Low ridges parallel to a river course. They are higher near the river and gradually slope away from it Panchayat. Local council Polder. Large, fully embanked area Rabi. The cropping period during the dry season (October through May) Regulator. A structure built to control water flow across an embankment at the head of a khal or a structure built in a river or khal to control water levels. Water is controlled by flap gates, vertical lift gates, fall boards, or a combination of these Stakeholder. An individual whose livelihood is directly affected by a water management system, be it positively or negatively Submersible embankment. An embankment whose crest level is designed below the highest normal river water level, typically occurring in the Northeast. These embankments delay monsoon flooding and protect the standing rice crop against flash floods Thana. The administrative unit of local government above the union level, consisting of 3 to 10 unions Union. The lowest unit of government in Bangladesh Union Parishad. Elected council at union level Upazila. Subdistrict Zamindar. Landlord Zila. District parishads References Abdel-Dayem, Safwat. 2000. 'Drainage experiences in arid and semi-arid regions.' In the "Eighth ICID International Drainage Workshop". New Delhi, India. pp.145-160. Abdur Rahman, K. 2002. Investment in land and Water in Bangladesh. RAP Publication 2002/09. Bangkok: Food and Agriculture Organization. Ahmed, Imtiaz, ed. 1999. Living with Flood: An Exercise in Alternatives. Dhaka: University Press Limited Asian Development Bank. 2002. "Jamuna-Meghna River Erosion Mitigation Project, Appraisal Mission Report and Revised Aide Memoire." Dhaka. ------. 2001. "Second Small-Scale Water Resources Development Sector Project, Appraisal Mission Report and Aide Memoire." Dhaka. BUET (Bangladesh University of Engineering Technology). 1999. "Macro Impact of Flood Control Projects on Aman Rice Production." Dhaka. Bangladesh, Ministry of Agriculture. http://www.bangladesh.gov.bd/moa/moa.html. (Accessed 2002). Bangladesh, Ministry of Environment and Forests. 1990. National Conservation Strategy of Bangladesh. National Conservation Strategy Secretariat, Bangladesh Agricultural Research Council, Dhaka. Bangladesh, Ministry of Water Resources. 1999a. Guidelines for Participatory Water Management, Dhaka. ------. 2001c. Draft National Water Management Plan. Water Resources Planning Organization (WARPO), Dhaka. www.warpo.org. ------. 2001b. National Water Policy. Web site of the Water Resources Planning Organization, Dhaka. www.warpo.org. ------. 2001. Bangladesh Water Act. Second draft, Dhaka. Bangladesh Water Development Board. 2002. "Coastal Embankment Rehabilitation Project 1995­2003." Information Brochure. Dhaka ------. 2001. Dampara Water Management Project, Canadian International Development Agency. Dhaka:CIDA. ------. 1998. Local Initiatives in Water Management: A Study of People's Water Management Practices in Rural Banglades., Technical Report No. 55, Systems Rehabilitation Project. Dhaka: Euroconsult, Haskoning, DHV, Bets for BWDB. ------. 1998. Water Management in Flood Control and Drainage Systems in Bangladesh. Vol. 1. Technical Report No. 50, Systems Rehabilitation Project. Dhaka: Euroconsult, Haskoning, DHV, Bets for BWDB. ------. 1994. Improved O&M in Water Management Projects. Technical Report No. 33, Systems Rehabilitation Project.Dhaka: Euroconsult, Haskoning, DHV, Bets for BWDB. 65 66 Controlling or Living with Floods in Bangladesh Brammer, H. 1996. The Geography of the soils of Bangladesh. Dhaka: University Press Limited Briscoe, J. 1998. "Report on Poverty Immersion Program--Fatehpur Village, Bangladesh." Office Memorandum. Washington, D.C.: World Bank. Chowdhurry, Alamgir, and Immerzeel, van W. 2001. "Flood Damage Rehabilitation of EIP Subprojects, Final Report." Commissioned by the Royal Netherlands Embassy, Dhaka. Unpublished. ------. 1999. Mid-term Review Mission, Small Scale Water Resources Development Sector Project.. Commissioned by the Royal Netherlands Embassy, Dhaka. Unpublished Compartmentalization Pilot Project, (Flood Action Plan 20). 1999. Interim Report. Draft. Dhaka: Flood Plan Coordination Organization. DFID (Department for International Development). 2001. From Flood to Scarcity: Re-defining the Water Debate in Bangladesh. University of Leeds Environment Center Working Papers Environment and Development Series No. 1. Leeds: DFID. Duyne, Jennifer. 1998. "LocalInitiatives for Sustainable Water Resource Management." Experiences of the BWDB SRP. Vol. 6. Dhaka: University Press Limited. Environmental and GIS Support Project for Water Sector Planning, Ministry of Water Resources. 1998. Environmental and Social Impact Assessment of Khulna-Jessore Drainage Rehabilitation Project. Dhaka:MWR. Ericksen N.J., Q.K. Ahmad, and A.R. Chowdhury. 1993. Socioeconomic Implications of Climate Change for Bangladesh. Briefing Document No. 4. Dhaka: Bangladesh Unnayan Parishad. FAO [Food and Agriculture Organization of the United Nations]. 1997. The State of Food and Agriculture. Rome: FAO. ------. 2002 Internet gateway to national information on land, water and plant nutrition, www.fao.org/landandwater/ ------. 2002 Aquastat: An information system on water in agriculture and rural development. Land and Water Development Division, www.fao.org/landandwater/ Government of the Netherlands. 2002. Integrated Planning for Sustainable Water Management (IPSWAM). Report of the Committee for Reformulation of the IPSWAM Program, Dhaka: Government of the Netherlands. Hossain, M. et al 2001. Changes in Agrarian Relations and Livelihoods in Rural Bangladesh, Insights from Repeat Village Studies. Dhaka: Agrarian Studies. Iqbal A. 1997. "The State of Land, Water and Plant Nutrition Resources in Bangladesh." Paper presented at the Food and Agricultural Organization's Regional Workshop on Land Vulnerability Assessment for Food Security using Agro-ecological Zoning/Land Resources Information System, Bangkok, November 3­8, 1997. Khan, T. A. et al. 2001. EIP Projects, an Assessment. Dhaka: University Press Limited. References 67 Knegt, J. 2000. Drainage in Developing Countrie:, A Review of Institutional Arrangements. CWP Research Paper No. 2. Wageningen University, Netherlands. Krug, J.A. 1957. Water and Power Development in East Pakistan. Report of a U.N. Technical Assistance Mission. New York, N.Y.: United Nations. June. Marandy, J. et al. 1993. Application of Rapid Rural Appraisal in the Preparation of Small Scale Water Sector Schemes in Bangladesh. Fifteenth Congress International Commission on Irrigation and Drainage The Hague, The Netherlands Md. Liakath Ali. 2002. An Integrated Approach for the Improvement of Flood Control and Drainage Schemes in the Coastal Belt of Bangladesh. Oliemans, W.J., and E. Kuindersma, E. 2002. Water Boards in Egypt: The Dutch Model? Nijmegen, Netherlands: International Commission for Irrigation and Drainage. Scheumann, Waltinia, and Claudia Friesen. 2001. "Seven Good Reasons for Drainage." In The Forgotten Factor: Drainage Its Role for Sustainable Agriculture. Bonn: German Development Institute. Wester, P., and J. Bron. 1998. Coping with Water. International Land Reclamation Institute Liquid Gold series No 4. Wageningen: Pudoc. Wilde, de K. 2000. Out of the Periphery: Development of Coastal Chars in Southeastern Bangladesh. Dhaka: University Press Limited. World Bank. 1998. "Water Resource Management in Bangladesh, Steps toward a New National Water Management Plan". World Bank Country Office, Dhaka, Bangladesh. ------. 2002. Considering Adaptation to Climate Change in the Sustainable Development of Bangladesh. Dhaka: World Bank, Bangladesh Office. WSIP [Water Sector Investment Program]. 1999. WSIP Draft Final Report. Dhaka: EPC Ltd. in association with Association for Social Advancement.