Show Summary Details

Page of

Printed from Oxford Research Encyclopedias, Natural Hazard Science. Under the terms of the licence agreement, an individual user may print out a single article for personal use (for details see Privacy Policy and Legal Notice).

date: 24 February 2021

Livelihoods in Bangladesh Floodplainsfree

  • Parvin SultanaParvin SultanaFlood Hazard Research Centre, Middlesex University School of Science and Technology
  •  and Paul ThompsonPaul ThompsonFlood Hazard Research Centre, Middlesex University School of Science and Technology


Floodplains are ecologically diverse and important sources of livelihood for rural people. Bangladesh is one of the most floodplain-dominated countries and supports the highest density of rural population in the world. The experience of Bangladesh in floodplain management efforts provides evidence, lessons, and insights on a range of debates and advances in the management of floodplain natural resources, the challenges of climate change, and the role of local communities in sustaining these resources and thereby their livelihoods. Although floodplain areas are primarily used for agriculture, the significance and value of wild common natural resources—mainly fish and aquatic plants—as sources of income and nutrition for floodplain inhabitants has been underrecognized in the past, particularly with respect to poorer households. For example, capture fisheries—a common resource—have been adversely impacted by the building of embankments and sluice gates and by the conversion of floodplains into aquaculture farms, which also exclude poor subsistence users from wetland resources. More generally, an overreliance on engineering “solutions” to flooding that focused on enabling more secure rice cultivation was criticized, particularly in the early 1990s during the Flood Action Plan, for being top down and for ignoring some of the most vulnerable people who live on islands in the braided main rivers. Coastal embankments have also been found to have longer term environmental impacts that undermine their performance because they constrain rivers, which silt up outside these polders, contributing, along with land shrinkage, to drainage congestion. Locals responded in an innovative way by breaking embankments to allow flood water and silt deposition in to regain relative land levels.

Since the early 1990s Bangladesh has adopted a more participatory approach to floodplain management, piloting and then expanding new approaches; these have provided lessons that can be more general applied within Asia and beyond. Participatory planning for water and natural resource management has also been adopted at the local level. Good practices have been developed to ensure that disadvantaged, poor stakeholders can articulate their views and find consensus with other local stakeholders. The management of smaller water-control projects (up to 1,000 ha) has been taken on by community organizations, and in larger water-control projects, there is collaborative management (also called “co-management”) among a hierarchy of groups and associations and the appropriate government agency. In fishery and wetland management, many areas have been managed by community organizations to sustainably restore common resources, although their rights to do this were lost in some cases. Associated with community management are successful experiments in adopting a more system-based approach, called “integrated floodplain management,” which balances the needs of agriculture and common natural resources, for example, by adopting crops with lower water demands that are resilient to less predictable rainfall and drier winters, and enable communities to preserve surface water for wild aquatic resources. Bangladesh also has had success in demonstrating the benefits of systematic learning among networks of community organizations, which enhances innovation and adaptation to the ever-changing environmental challenges in floodplains.


Since the early 1990s research and development in the Bangladesh floodplains have given rise to debates, evidence, knowledge, and lessons that have wide international relevance, across several fields, to the challenge of securing and improving livelihoods of the inhabitants of rural floodplains. After introducing the main characteristics of Bangladesh floodplains, this article reviews the debates over the impacts of structural approaches to flood management and public participation in water management at the time of the Flood Action Plan. This is followed by a summary of knowledge on the significance and value of floodplain natural resources, particularly commons, in rural livelihoods of Bangladesh. Those sections set the scene and context for the following two main sections that summarize knowledge and recent research: first, on collaborative management, community-based governance and participatory planning in floodplains; and second, on adaptation and integrated resource-management approaches to the challenges of changing environment and climate. The conclusion highlights some of the contributions to knowledge discussed, gaps, and remaining research questions.


Global Floodplains

A floodplain is made up of flat or nearly flat land that is adjacent to a stream or river and experiences occasional or periodic flooding. It includes the floodway, which consists of the river channel and adjacent areas that carry flood flows, and also the fringing areas, which are covered by floods but do not experience a strong current. Flood waters carry suspended sediment that has been eroded from upstream areas and deposit this as silt on the floodplain. It is widely believed that the sediments or silt deposited by rivers to build up floodplains and deltas are rich in nutrients, and thus form the basis for productive ecosystems and agriculture. Yet even this perceived wisdom—that silt is the main source of nutrients—is a contested simplification of a complex system. Nevertheless, floodplains comprise some of the richest ecological systems on the planet (Ramsar, 2001) and can support highly productive agriculture when they are not flooded. Human interventions to change their hydrological regime began early (Hillel, 1992), and in the early 21st century floodplains are among the most densely populated areas on the planet. Floodplains provide rich ecosystems and diverse ecosystem services that have supported rural communities for generations. Floodplains are particularly complex systems in terms of biophysical, socioeconomic, institutional, cultural, and political dimensions, but modern development interventions have frequently failed to account for this complexity.

Although all rivers have floodplains, some floodplains are of particular note for their scale or social-ecological systems. Examples of these in the Americas include the Mississippi River floodplain in the United States, with its extensive modifications and embankments; the diverse grass- and marshland-dominated Pantanal floodplain of Brazil, Bolivia, and Paraguay; and the flooded forests of the largest floodplain along the Amazon River in Brazil. Africa is notable for inland floodplain deltas, such as the Okavango in Botswana, and for seasonal floodplains that provide an important source of pastoral livelihoods, as well as for agriculture and fisheries in the Sahel. Many of Asia’s floodplains are cradles of civilization that have been heavily modified over millennia, for example, the combination of a heavy sediment load and embankments present along its length for centuries has resulted in a river bed raised higher than its floodplain along the Yangtze River.

Bangladesh Floodplains

Bangladesh is a floodplain deltaic country formed by the deposits of the three major river systems of the Ganges, Brahmaputra-Jamuna, and Meghna Rivers (Figure 1). Bangladesh is one of the most floodplain-dominated countries in the world and also one of the most densely populated, at over 1,100 people per km2. Including tributaries and distributaries there are around 700 rivers in Bangladesh stretching over 24,140 km, thousands of smaller channels, floodplain depressions (known as beels), and extensive seasonally flooded lands that collectively form the floodplain ecosystems (Akonda, 1989). Estimates of the area of floodplain in Bangladesh range up to 80% (Brammer, 1990), and about 25% to 33% of the country remains under water every year for four to six months during the monsoon (rainy season).

Figure 1. Map of Bangladesh showing significant floodplain locations and place names.

Floods, from slow-rising main rivers, flashy smaller rivers, heavy rainfall, and tidal or storm surges, are a recurring phenomenon. Rural life is well adapted to the normal monsoonal patterns of inundation, known as “borsha”, but if the frequency, timing, intensity, or duration of inundation is atypical and falls outside the usual range, then it is termed a “bonna” (flood) and becomes a serious threat to lives and livelihoods, particularly for the poor floodplain inhabitants. Strong river currents erode croplands and even destroy or damage homes and villages. By the same token, unusually low rainfall, including untimely rains and extended dry seasons or periods of drought, create water stress that negatively impacts agriculture and aquatic life in the floodplains. In addition to floods, erosion, dry-season water stresses, and salinity intrusion (seen in the southern parts of Bangladesh) are the most frequently experienced water-related hazards and stresses in Bangladesh floodplains, as found in a wide-ranging study (Flood Hazard Research Center Bangladesh, 2013).

The Bangladesh floodplains are fed by various sources of water (rivers, groundwater, rainfall, and tidal water) and many of them have more than one water source. The contribution of each water source and its impact on livelihoods can change on a seasonal or annual basis, depending on water availability and changes in the water regime. Rains or flash floods that come earlier than normal can damage dry-season crops just before harvest. Even if farmers are able to harvest their crops at such times, there will be a lack of places for drying crops, particularly paddy (rice); moreover, dry land for storing rice straw (used for livestock fodder) becomes scarce. In the regions near the hills on the eastern side of the country, the damage caused by flash floods (despite their short duration) can be more severe than that caused by slower-rising and longer-duration riverine floods because of sudden flooding without warning, high water velocity, and debris load. Nevertheless, it is the large-scale main river floods that draw international attention and with their widescale impact were a significant setback to economic growth in the past when the national economy was heavily agriculture dependent (Benson & Clay, 2002). Since 1980, major floods occurred in Bangladesh in 1987, 1988, 1998, 2004, and 2007. Estimates of the numbers of people affected by this flooding vary considerably. The 1988 flood was a 1-in-100-year event that inundated 42% to 57% of the country (Rashid & Pramanik, 1993; Miah, 1988). Over 2,000 people died; 1,990 km of embankments were damaged; and total losses, including infrastructure and crop damage, were about US$1.3 billion (World Bank, 1990). In 1998 about 60% of the country was flooded, with similar impacts—about 1,000 people died, about 2,000 km of embankments damaged, and losses were estimated at up to US$2.8 billion (Brammer, 2004; Hossain, 2006). However, the relative significance of such major floods has declined over time: the potential reduction in gross domestic product associated with major disasters has gradually become smaller as the national economy has grown and become less dominated by agriculture (Benson & Clay, 2002) and as disaster preparedness, including warning systems, have improved.

The 2011 census conducted by the Bangladesh Bureau of Statistics suggests that up to 120 million people lived in the country’s floodplains at that time (see Figure 2). Floodplains provide a range of ecosystem services. A large proportion of the rural poor depend on natural water bodies in the floodplains for their livelihoods. Their subsistence is based on food production, fishing, harvesting wetland plants, plying country boats, and other activities that depend on healthy aquatic ecosystems.

Figure 2. A typical seasonal floodplain during the monsoon, Narail district, southwest Bangladesh.

Photo by Paul Thompson.

Floodplain fisheries have been of particular significance for generations, and about 11 million people were estimated to depend directly or indirectly for their livelihoods on the Bangladesh fishery sector in 1990 (World Bank, 1991). Some 82% of the households that depend on fishing for income are poor (WorldFish Center, 2003). Up to 80% of rural people and about half of rural poor households living in the floodplains catch fish and use other aquatic resources, and up to 70% of animal protein consumption in Bangladesh is derived from fish (Minkin, Rahman, & Halder, 1997; Muir, 2003; Toufique & Gregory, 2008). Catches from these fisheries, which may be termed common-pool, capture, or open-water fisheries, have been declining. Fish consumption fell by 11% between 1995 and 2000 (and by 38% for the poorest households), and it is estimated that inland capture-fishery catches fell by 38% between 1995 and 2002 (Muir, 2003; however, aquaculture production has risen rapidly; see the section “Commons and Enclosure”). Bangladesh’s floodplain wetlands support over 260 fish species (Rahman, 2005). Although there has been no known extirpation in Bangladesh, 54 (21%) of Bangladesh’s freshwater fish species were considered to be threatened with national extinction in 2000 (International Union for the Conservation of Nature [IUCN] Bangladesh, 2000), and 64 (25%) of these species were considered to be threatened with national extinction in 2015 (IUCN Bangladesh, 2015).

Besides crops and fish, the main provisioning services (i.e., resources directly used by people) in the floodplains are wild plants (aquatic grasses and other plants that are used for grazing and livestock fodder, construction, fuel, and human food), snails (used as feed in duck and fish farms), and minor resources, such as crabs.

Globally, most floodplains have been modified, to a greater or lesser extent, by human interventions as diverse as small-scale agriculture; expansion of villages and towns, constructing bridges; and forest clearing in catchments. However, one of the foundations of ancient and modern civilizations in floodplains, the raising of earthen embankments, also known as “dikes” or “levees,” to change patterns of flooding has a special impact on the livelihood potential in floodplains. This is also true in Bangladesh, where there is a long history of local water-management earthworks. Moreover, since the 1960s, water-resource management to control flooding in Bangladesh has largely relied on a centralized engineering approach.

The normal annual flooding in Bangladesh is considered to provide numerous benefits, such as giving Bangladeshis common access to the large, natural, floodplain fishery; depositing fertile loam on agricultural fields; and flushing stagnant water in low-lying areas. However, the second of these, the common view that floods bring fertile silt to floodplains, appears to apply to only part of the system in Bangladesh—the active floodplain and braided river channels. Brammer (2004), who has pointed out that silt has a low organic content and documented shallow soils in much of the floodplain, argues instead that high river levels (which result in silt being transported into the Bay of Bengal) trap rainwater, which is the source of normal flooding, in the rest of the floodplain. Floodplain soil fertility is not the result of silt, but because this clearer rain-derived water supports blue-green algae, which contribute significant amounts of nitrogen to the floodplain system, in addition bacteria add to soil nutrients when soil is first flooded, and thirdly leaves decompose during flooding, all of which contribute to fertile conditions supporting crop growth (Brammer, 2004). Although the Environment and GIS Support Project for Water Sector Planning report [EGIS], 1997) revealed that shallow flooded lands support relatively more wild fish in terms of biomass and diversity than deeper flooded lands and that the heavily silt-laden Jamuna River is of modest importance for fisheries, there is a lack of information comparing fish diversity and catches between clearer and more turbid flood waters.

Local communities have a direct interest in maintaining the floodplain ecosystem services that their livelihoods depend on. However, the complex, dynamic nature of the systems makes managing them a challenge. Bangladesh wetlands have ample water for half the year in the wet season; but the limited amount of surface water in the dry season drives productivity. In the dry season, the surface water that is naturally left in river channels and beels is needed for agriculture, domestic use, and the survival of fish. Agricultural development has largely focused on rice, and much of the rapid growth in rice production since the 1980s has depended on dry-season irrigation of high-yield rice varieties initially pumped in from surface water, in the 1970s, and then, beginning in the 1980s, this irrigation was done using groundwater (which has lowered the water table; Sultana, Johnson, & Thompson, 2008; Brammer, 2010). Embankments and flood control play an important part in this by protecting rice crops from flooding in the maturing and harvesting periods, while sluices have helped to drain out water from floodplains and beels in the post-monsooon. Despite changes in national policies that call for an end to the draining of remaining wetlands (Ministry of Water Resources, 1999), floodplains continue to be converted to sites of agriculture, aquaculture, and urban-related uses.

Issues Raised by the Flood Action Plan

Much of the debate in Bangladesh over how best to enhance the livelihoods of floodplain inhabitants and reduce their vulnerability was catalyzed in the Flood Action Plan of Bangladesh (hereafter, FAP). In response to severe floods in 1987 and 1988 reviews and recommendations for flood and water management in Bangladesh were made by France, Japan, United States, and the United Nations Development Programme [UNDP]. In an attempt to use detailed evidence to consolidate the very divergent views in these reviews, and to develop a comprehensive plan to manage flooding and reduce losses, a consortium of 17 donors responded to requests from the Government of Bangladesh by undertaking the FAP which involved 26 studies and pilot projects undertaken during 1990–1995. FAP created wide-ranging debate and controversy (Brammer, 2000), the initial underlying concept was that the main rivers could be fully embanked as a solution to flooding (World Bank, 1990). From the beginning, it drew criticism as a megaproject that would change the socio-ecological character of Bangladesh (Boyce, 1990). With the benefit of hindsight, several authors have reflected on the FAP, its legacy and continued relevance; these reflections can be looked at together with the findings and debates during FAP.

Prior to FAP from the 1960s up to late 1980s an engineering coalition or paradigm of structural responses to floods had dominated government and international donor investments in Bangladesh floodplains (Sultana et al., 2008). During the early 1990s, nongovernmental organizations (NGOs) and civil society groups publicly questioned the FAP under the newly established parliamentary democracy after the overthrow of General Ershad’s regime in late 1990. Debate centered on the past and future impact of embankments in Bangladesh in terms of technical performance and feasibility, maintenance, economic returns, negative effects on the environment and fisheries, and inequitable impacts that disadvantaged poorer people.

Impacts of Embankments on Fisheries

At the time of FAP about 7,550 km of flood-control embankments had been built across Bangladesh, including the coastal zone and along main, secondary, and minor rivers. Sluices (water-control structures; see Figure 4) of various sizes were constructed in almost all of these embankments to help drain the water, and also to let water in when needed and possible (see Figure 3). Locally, the operation of sluices has contributed to draining floodplain wetlands for the benefit of agriculture, but embankments had negative impacts on fisheries because they caused a disruption of migration and the loss of wetlands (Ali, 1990, 1997; Halls, 1998; Hughes, Adnan, & Dalal-Clayton, 1994). Many fish species follow annual cycles according to the rise and fall of normal floods. Many smaller species travel short-distances from beels to the floodplain to breed and feed; other (often larger) species migrate along the rivers to spawn, and their fry and spawn disperse onto the floodplain to feed (Sultana & Thompson, 1997). Thompson and Sultana (2000) found that fishers reported negative impacts in 12 out of 18 flood-control projects studied. On average, the reported daily catch per person declined 60% from the pre-project levels, compared with a decline of 74% from previous levels in control areas outside the same projects. Studies in the Tangail pilot compartment revealed the strong seasonality of floodplain fisheries, the impacts of intense fishing pressure (which increases at sluices, where fish are concentrated), decline in both catch volume and diversity that would be associated with flood control, and the mortality of fish hatchlings passing through sluice gates (de Graaf, Born, Uddin, & Marttin, 2001). The FAP was notable for bringing to the public attention the negative environmental impacts of flood-control works, along with the significance of declining capture fisheries and the nutritional value of fish, including the relative advantages of sustaining the supply of the wild small fish that are caught and eaten by poor floodplain households. Because these fish are eaten whole, they are important sources of micronutrients, containing higher levels of not just calcium but also vitamin A and iron compared with cultured species (Thilsted, Roos, & Hasan, 1997; Thompson, Roos, Sultana, & Thilsted, 2002).

Figure 3. Women and children collecting aquatic plants (Singra or water chestnut) for food in Hakaluki Haor, northeast Bangladesh.

Photo by Paul Thompson.

Active Floodplain and Chars

Given the FAP’s focus on structural flood mitigation, there was concern about the fate of the people living in the active floodplain, including in the chars—the silty-sandy lands and islands within the braided main rivers. These areas were already the most flood prone and could be subjected to worse flooding if stronger complete embankments were built along these rivers. A char is defined as a “mid-channel island that periodically emerges from the riverbed as a result of accretion” (Elahi, Ahmed, & Mafizuddin, 1991). Socially and ecologically, the accreted riverine lands that link to the mainland in the dry season were also considered to be chars under the FAP (FAP 16/19, 1993) (see Figure 4). Chars are inherently hazardous places to live. Public policies might be expected to control settlement in hazardous locations, but there are no such planning regulations in Bangladesh; moreover, erosion and shifting main river courses mean that people on the mainland also lose their land to the rivers, and then they wait for “their” land to reappear when a char accretes in the same area. To adapt to these changing conditions, many char people have been forced to move home several times during their lifetimes (even multiple times a year) and to make use of nearby embankments as important refuges when homes are eroded, and to seasonally migrate to towns for work. The communities living in these chars had largely been neglected in terms of the provision of public services and development projects, and obviously (since they lie within the active river channels) fell outside of water-management investments and benefits, but were also treated as socially distinct and inferior to mainland people.

Figure 4. A typical sluice gate in northwest Bangladesh.

Photo by Paul Thompson.

Under the FAP an inventory and related studies focused on these areas (FAP 16/19, 1993) highlighting their dynamic character because of frequent erosion and accretion. For example, the Jamuna River widened by 128 meters each year from 1973 to 2000 (EGIS, 2000) because mainland along the banks eroded. Although char islands appeared within the main river they have limited productivity. These chars become vegetated in one to two years and are then settled and cultivated despite the hazards and livelihood insecurity, which mostly takes another two years (EGIS, 2000). Dry-season crops such as pulses, millets, and groundnuts are major income sources for the people living on the chars; cattle grazing is also common, and grasses are cut for thatching and construction. In 1993, an estimated 0.6 million people lived in the riverine chars and were arguably the most affected by major floods (EGIS, 2000). By highlighting the vulnerability of the livelihoods of the char dwellers, the FAP left a legacy that was seen decade later, when development projects began focusing on “nonstructural” approaches to mitigating the losses from flooding—for example, flood proofing by raising the plinths (floor levels) of houses, and transferring assets to poor households such as cattle to improve income-earning potential.

Debate over FAP

During the FAP there was considerable debate over who should participate in water-management planning and implementation. One of the guiding principles of the plan was that it should “encourage popular support by involving beneficiaries in the planning, design and operation of flood control and drainage works” (World Bank, 1990). However, a government task force, in early 1991, concluded that public-awareness programs had not been incorporated into the plan (Report of the Task Forces on Bangladesh Development Strategies for the 1990s, 1991). Other criticisms of flood control emerged as part of the FAP debate. For example, flood-control projects involve acquiring substantial amounts of land to use for embankment construction. Problems of inadequate compensation to the owners of that land and the procedural complexity of the land-acquisition process were found to cause major economic and social suffering to those who lost their land as a result of it, especially small farmers, who lose their crop land, and households, which lose the land where their homes are located. Ultimately, these problems are rooted in poor governance, which in Bangladesh is arguably at its most compromised in land administration, where land acquisition for flagship projects is still found to be inequitable (Atahar, 2013), and where reviews commissioned by the government find that the present system enables widespread fraud and malpractice (Hossain, 2015).

Civil society questioned the legitimacy of decision-making under the plan, especially in those components of FAP that were based on predefined interventions—notably, the concept of “compartmentalization,” whereby the lands behind a continuous line of embankments along the main rivers would be protected by secondary embankments to form a series of compartments that could be selectively opened to let in floodwater, relieving pressure on other areas). The combination of government and donor pressure to take action resulted in decisions to construct two pilot compartments, but this attracted local opposition and resulted in demonstrations and lawsuits brought by environmental NGOs wanting to halt the project and compensate local people who reported or claimed adverse impacts from embankment and sluice construction. Although the debate was focused on flood infrastructure, it can be seen as part of the greater challenge of finding a balance when economic development and infrastructure bring wide benefits but also result in big losers, who in this case tended to be poor, for whom there was no effective compensation mechanism. Nevertheless, adoption of participatory approaches in the water sector (see the section “Participation in Water Management”) can be traced to Flood Action Plan.

Significance of Floodplain Natural Resources to Livelihoods

Approaches and simplifications to livelihood analysis that assume undifferentiated communities have been criticized (Mehta et al., 1999). Bangladesh floodplain inhabitants pursue diverse livelihood strategies that result in competition and conflict, but also the potential to cooperate over access and the use of natural resources (Barr et al., 2000).

Floodplains are complex systems. Private land becomes a seasonal commons when it is flooded, and people make use of a multitude of natural resources, all connected through water. Much of the floodplain land is de jure or de facto privately owned and farmed, and water stands there for up to six months. The beels, which hold water year round, and rivers are public or state property. The government leases fishing rights in the beels to the highest bidder, but it has made the use of rivers open access. However, when floodplains are inundated during the wet monsoon season, local people have customary rights to catch fish and collect other aquatic resources (such as plants, used food, fodder, fuel, and building materials, and snails, used as feed for ducks and in shrimp farming) from both public and private lands. Floodplains are vital to capture fisheries, where fish breed and grow out during inundation before moving to deeper areas for the dry season (Shankar, Halls, & Barr, 2004). This means that common-pool resources exist during the wet season in areas that are privately owned (Sultana & Thompson, 2008). Access to many floodplain natural resources is not clearly defined legally, but the poor have access through custom, local tradition, negotiation, or conflict. The better-off landowners have the right to privatize these resources and prohibit their use by others, but this is only practicable in the dry season. When water levels fall, landowners are able to exclude others from their lands, harvest whatever fish and aquatic plants remain, and then cultivate crops. However, from the perspective of the poor, the aquatic resources present on this land, or in the water above this land when it is flooded, are a common-pool resource. This leads to tensions, for example, when water levels are falling. Moreover, use of floodplain resources is differentiated by gender—it is rare for women to fish for a livelihood, although in some areas poor women and children do catch fish for subsistence, and women collect other aquatic resources (Deba, Haque, & Thompson, 2014; Sultana & Thompson, 2008). Government interventions often ignore these complexities (see Figure 5).

Figure 5. Fishing using a beach seine net in the Jamuna chars.

Photo by Paul Thompson.

Since the 1990s, understanding of the significance of commons and of the local institutions that regulate their use has grown rapidly, inspired by research led by Ostrom (1990), Berkes and Farvar (1989), and others. The Bangladesh experience adds to this knowledge by highlighting the complexity of floodplain commons in terms of ecological linkages, human use, and institutions encompassing a range of property rights. It confirms that property rights are more complex than being simply private or public and that different types of rights may exist for the same physical area in different seasons and for different components of the suite of natural resources present in that area (see also analysis by Agrawal & Ostrom, 2001; Ostrom & Schlager, 1996; Meinzen-Dick & Gregorio, 2004).

Value of Floodplain Natural Resources

There have been few economic assessments of the value of floodplains in Bangladesh. Thompson and Colavito (2007) valued uses of Hail Haor in 2000—haors are large saucer-shaped depressions in northeast Bangladesh that in the monsoon are entirely deeply flooded, but in the dry season become a mosaic of permanent water in beels, marshy areas, waterways, and dry land. At the time, Hail Haor was being overexploited, but the authors estimated that the annual economic value generated from the 12,300 ha that were flooded in 1999 was equivalent to US$7.98 million (almost US$650 per ha). The main human-use value from the site comprised commercial fisheries, 12%; subsistence fisheries, 18%; aquatic plants used by local residents and by tea estates, 28%; dry-season rice, 14%; biodiversity (spending on conservation used as a proxy), 10%; pasture, 9%; flood-control services, 5%; recreation, 2%; and transportation, 2%. Overall, 67% of the annual value of the haor came from wild natural resources. By 2004–2006, following conservation-based management improvements, the fish catch estimated over the whole haor had increased to 322 kg/ha of wetland, an 88% increase over the 1999 baseline. This yielded, at constant prices, an estimate of the value of direct uses of the haor and its products of US$10.9 million a year (57% derived from fish). High yields of fish were sustained here in 2011, 2012, and 2014, averaging 390 kg/ha. These benefits extend beyond the 3,770 households that fish to earn a living, as many more people catch fish and use aquatic plants for their subsistence.

A similar study (IUCN Bangladesh, 2006) of another large floodplain wetland—Hakaluki Haor—found an economic value of about US$414 per ha (reworked to correct calculations in that report). It estimated that 80% of local households used wild natural resources and confirmed the economic significance of noncrop common resources (75% of use value) compared with rice (which contributed only 25% of the use value). Fish and grazing were the main uses.

Commons and Enclosure

A study by Sultana (2012) investigated a new trend in Bangladesh since the 1990s: the private enclosure of floodplains for aquaculture. Ostensibly, aquaculture use retains their wetland floodplain status; but a number of adverse impacts on poor people, aquatic resources, and the wider floodplain system were revealed. First, the trend of converting seasonal commons to private enterprise use is rapid and widespread: in three study areas with different types of floodplain aquaculture (small individual operations, small group enterprises, and larger enclosures built by local companies) over 500 aquaculture enclosures were documented, and the area of aquaculture increased by 30% to100% annually between 1990 and 2008 (Sultana, 2012). Similar growth was found in a fourth more deeply flooded area arising from capture of lands by rich investors (Table 1).

Table 1 Growth of area of aquaculture enclosures (in hectares) in four study areas





Hail Haor





















Source: Narail, Gazipur, Comilla (Sultana, 2012, reworked data); Hail Haor, unpublished data from the Climate Resilient Ecosystems and Livelihoods project, Winrock International, Dhaka, Bangladesh.

These systems are productive and profitable for the entrepreneurs, although costs are high (Mustafa & Brooks, 2009). Toufique and Gregory (2008) investigated the so-called “Daudkandi approach” (to floodplain aquaculture (where local people in Daudkandi area of Comilla District started to form companies to invest in building bunds around large areas of floodplain to both retain water and keep out floods, so that they could stock the area with fish without fear of the fish escaping). In two villages these authors found qualitative evidence that the benefits of floodplain aquaculture tend to accrue to better-off people, with elite capture of the boards of the companies formed for this enterprise. However, catches of wild fish were reported to have declined to between 44% and 58% of the pre-enclosure levels, and fish diversity also fell. These mostly small wild floodplain fish were caught for food by poor people who are excluded from the aquaculture enclosures, and there was also a loss of aquatic plants, snails, and grazing. All of this mainly impacted on the poor—landless men and women, fishers, and marginal farmers (Sultana, 2012).

Participation and Community-Based Management


Floodplain natural resource management in most developing countries has traditionally been top-down and sectoral, led by technical experts, and resulting in low compliance with rules set centrally, limited local contributions to running costs, and plans and designs that do not take into account local circumstances. This led to pressure from NGOs for change. Internationally, a change in approach was signaled by Principle 10 of the Rio Declaration on Environment and Development (United Nations Conference on Environment and Development 1992), which states that environmental issues are best handled with the participation of all concerned citizens “at the relevant level” and thereby recognized that the top-down approach was not working and that citizen participation should be encouraged. In floodplain management, the same level of participation is not realistic of all concerned parties. A people-centered approach in which users and government agencies make decisions jointly or when community organizations lead is more difficult when there are larger and more complex natural resource systems and management issues. Participation is often intended to empower the poor. But simply having meetings where all can attend is insufficient to enable local people to take responsibilities, as poor people tend not to be organized to take collective action, Moreover experience at the local level may be insufficient to give local people broader perspective on the scope and complexity of decisions in larger systems unless communities can cooperate with each other and with other experts.

Community-Based Natural Resource Management

Community-based management refers to systems in which local resource users have the main rights and responsibilities for management decisions and actions for a resource or ecosystem rather than government. This may arise either from government recognizing traditional local institutions, or government devolving powers that had been more top-down and centralized to the newly formed community organizations. Community-based management is thought to have several advantages over government-led or top-down management, particularly for common-pool resources (where it is difficult to exclude users, but there is subtractability—for example, people sell or eat the fish they catch—which overcomes the “tragedy” of open access in commons (Hardin, 1968). Looking at the management of fisheries, Pinkerton (1989) identified six benefits of greater community participation:


Users cooperate in planning to enhance or conserve natural resources (sustainability).


Users share the costs and benefits of improved management (economic equity).


It improves conflict resolution among users (intragroup social equity).


The position of users when dealing with other stakeholders is enhanced by their being organized (inter-group social equity).


Users and government share data and their understandings.


Security of tenure for local users over natural resources improves trust between users and government, consequently:


users have an incentive to take a longer term perspective (sustainability), and


enforcement of rules improves as these rules are set by users so compliance improves, reducing “transaction costs” (efficiency).

As part of the worldwide trend toward community-based approaches, from the 1990s onward community-based management has been tested, promoted, and expanded in Bangladesh. Community-based management has been promoted as a means to reduce poverty, conserve natural resources, empower local communities, and promote good governance. Given limited traditional institutions (rules and related organizations regarding access to water or fish for example) in the Bangladesh floodplains, development of community-based organizations (CBOs) for resource management is necessary, in particular to improve and sustain benefits to the poor. CBOs are generally grass-roots organizations of direct local natural-resource users (excluding middlemen or elites, but in some cases including local opinion leaders who champion sustainable practices and a greater role in decision making for the poor; they are formally recognized by the government as having responsibilities for an area, but no government officials are involved as members). Without these formal CBOs, there is a risk that benefits from management of floodplain resources will be lost or more easily captured by local elites.

Community-based management has created opportunities to empower communities, and especially poor resource users. But involving the whole community when the resource is valuable or the CBO has a large budget tends to attract local elites seeking profits. For example, CBOs formed with all villagers in a fisheries project that offered substantial public investment in stocking fish led to domination by elites seeking to profit from subsidies and the approach had to be revised to reform those CBOs to only comprise of fishers (Fourth Fisheries Project, 2005). Empowering the poor for livelihood improvement in floodplains involves capacity building; targeting the disadvantaged, including women; and linking participants with services.

Collaborative Management (a.k.a. Co-management)

In community-based management, the CBO has the majority of the powers and responsibilities. Distinct from this is the collaborative management approach (also called “co-management”). Definitions of co-management include “the sharing of power and responsibility between the government and local resource users” (Berkes, George, & Preston, 1991, p. 12), and “a situation in which two or more social actors negotiate, define and guarantee among themselves a fair sharing of the management functions, entitlements and responsibilities for a given territory, area or set of natural resources” (Borrini-Feyerabend, Taghi Farvar, Nguinguiri, & Ndangang, 2000, p. 1).

In some of the larger floodplains in Bangladesh (for example, in water-management polders and in the Hail Haor wetlands), stakeholder-participation is achieved through a multitier institutional arrangement in which a committee of government and CBO representatives share a coordinating and oversight role over CBOs which take responsibility for management of local areas and resources within a co-management framework (Sultana & Thompson, 2010). In addition, in all cases the CBOs are legal entities registered as social welfare organizations or as cooperateves with the government, and formed under the sponsorship and advice of a government agency, so they have the prospect of advice and oversight from government, although in practice, this depends on the interest of the concerned officials.

Participation in Water Management

In the context of problems associated with past flood-control projects (Thompson & Sultana, 2000; see also the earlier discussion on Flood Action Plan (FAP), and experience from rehabilitation of large flood-control works (Soussan & Datta, 1998), guidelines on public participation for the water sector were first prepared under FAP (Hanchett, 1997). The 1999 National Water Policy called for inclusive water management, taking into consideration the national goal of poverty alleviation. Participation guidelines have since evolved, bringing together initiatives from two government agencies—the Bangladesh Water Development Board (BWDB) and Local Government Engineering Department (LGED). A general guideline was formalized for all publicly funded water-resource projects that brought together good practice from both concerned agencies (Ministry of Water Resources, 2001). It aims to ensure local stakeholder participation in identifying of problems, feasibility studies, detailed planning, implementation, operation and maintenance, and monitoring and evaluation. However, it does not enable of participation at the higher levels of agenda setting or developing strategies or polices. Among its main concerns are the long-term institutional arrangements for water and floodplain management. It stipulates that flood-control projects covering up to 1,000 ha (constructed under the LGED) are owned by local government and managed by the communities; those of 1,000 to 5,000 ha (constructed by the BWDB) are owned by government agencies and managed by community organizations together with local governments, and those above 5,000 ha (also constructed by the BWDB) are owned by government agencies but may be co-managed by devolving some powers and responsibilities to CBOs (water user groups).

The number of CBOs rapidly increased under this framework. LGED registered 463 cooperatives to manage floodplain water infrastructure by 2003 created under the Small-Scale Water Resources Development Sector Project, and 600 or more were developed under various subsequent initiatives. Under the BWDB’s multitier model of co-management for polders and projects, local water-management groups are federated into associations; for example, the Blue Gold Project, in 2016, worked with 339 water management groups and 31 associations in 14 coastal polders. These multitier co-management systems are in many cases not (yet) working as intended. For example, Bernier, Sultana, Bell, and Ringler (2016) found that water timing and release often depend on local elites (and may require payments), which can reduce the amount of water available locally in the dry season if it is diverted for their interests, at the cost of crops or fisheries in other parts of the system.

Management of Fisheries and Environment

Meanwhile, policies that were supposed to reserve waterbody leases for poor minority fishers have been abused; powerful local people often obtain leases in the name of fisher cooperatives. The failure of this long-established, revenue-based administration of inland floodplain fisheries to secure fisher livelihoods, along with awareness of international initiatives in community-based natural resource management, led to several donor-supported projects involving NGOs and the Department of Fisheries that established community-based fisheries management. Initial efforts demonstrated that natural fishery productivity can recover when silted-up channels between floodplain wetlands and main rivers are re-excavated (Rahman, Capistrano, Minkin, Islam, & Halder, 1999) and focused on organizing traditional (Hindu) minority fishers to manage fisheries, but only assured them access for three years (Thompson, Sultana, & Islam, 2003). Donor interest was complemented by mutual benefits between the Department of Fisheries and fishing communities. Neither the department nor fishers had secure access or decision-making roles in waterbodies administered by the Ministry of Land, but if the department cooperated with donors, fishers, and NGOs, it could have waterbodies reserved for CBO management, under department oversight, and gain a greater role in the management of fisheries. Formalized recognition and alliance with the Department of Fisheries offered these communities access to fisheries through ten-year agreements reserving these rights, brokered by the Ministry of Fisheries and Livestock and the Ministry of Land. A series of projects from about 1994 onward (supported by the Ford Foundation, UK Aid, the World Bank, USAID, the International Fund for Agricultural Development, and the Government of Bangladesh) helped fishers to organize in individual waterbodies, and, in some cases, to coordinate these activities with the government for larger wetland areas which contain multiple waterbodies. This resulted in the formation of about 300 fisheries-management CBOs by the late 2000s and the introduction of local conservation measures, such as fish sanctuaries and closed seasons (Sultana & Thompson, 2012).

A separate co-management approach has developed in wetlands that have been declared ecologically critical areas (ECAs). Under the Bangladesh Environmental Conservation Act 1995, eight ECAs were, in 1999, declared threatened or degraded ecosystems, mostly wetlands. Typically, the ECAs, such as Hakaluki Haor (a large floodplain in the northeast), comprise a mix of public lands (waterbodies and other public lands) and private lands. In the ECAs for which project funding was available, the model adopted was to form village conservation groups as local cooperatives, coordinated by tiers of government committees. Most of these groups do not have specific rights over defined areas, but a few groups in Hakaluki Haor have been given the responsibility to protect waterbodies that have been set aside as sanctuaries.

Participatory Planning

One notable methodology that has been developed in Bangladesh for building consensus among diverse stakeholders for sustainable management of natural resources is participatory action plan development (PAPD). It is a process that takes individual and community concerns into account. Participatory approaches have been criticized as a kind of “tyranny” that, for example, reinforces existing social relations or fails to understand local power relations, or has focused on tools rather than empowerment (Cooke & Kothari, 2002; Holmes & Scoones, 2000; Mosse, 2002). PAPD is designed to ensure that poor people’s interests are voiced and represented on an equal footing with those of more powerful stakeholders (Sultana & Thompson, 2004).

The principle behind PAPD is that members of any stakeholder category, but especially the disadvantaged, are better able to express their views separately from other (dominant) categories of people. However, separate workshops will fail to develop a shared understanding of common problems and possible win-win solutions. PAPD seeks consensus among the stakeholders on actions that are needed to address common problems. It encourages participants to respect one another’s concerns by having two rounds of divergent and convergent sessions. Since the early 2000s it has been applied in a wide range of projects and locations in Bangladesh. The key steps in a full PAPD are


Problem census (with each stakeholder group separately).


Compilation of stakeholder problem rankings by facilitators.


Plenary with stakeholders and local leaders (to review and agree on main problems for solution analysis).


Solution and impact analysis (with each stakeholder group separately).


Plenary with stakeholders and secondary stakeholders (to present the process, identify feasible solutions, discuss institutional arrangements and next steps).

PAPD actively encourages participation by the poor, empowers disadvantaged stakeholders, and creates scope to evolve an internally driven agenda by ensuring all have a voice. Sultana and Abeyasekera (2008) studied 36 floodplain fisheries CBOs, half of which were initiated through PAPD and half without PAPD. They found statistically significant differences associated with PAPD, concluding that PAPD led to more efficient formation of CBOs, and saved time and resources in the process leading to actions. Moreover, CBOs that started with PAPD took up more conservation related interventions and faced fewer conflicts than CBOs that did not benefit from PAPD, indicating a greater level of consensus and trust among stakeholders following PAPD. Ultimately, this was associated with benefits for poorer fishers in terms of fish catches and in terms of local social status.


Despite these achievements, participatory floodplain resource management in Bangladesh continues to face challenges:

Biophysical environments are complex—floodplain systems tend to spread across village boundaries, so that different communities are involved in decision-making and management, as well as across administrative and sectoral boundaries which complicates their governance.

Socioeconomic environments are complex, with many different stakeholder groups each with different seasonally shifting livelihoods needs.

Lack of coordination among different government agencies each taking up activities in the same area.

Weak and uneven commitment of government agencies to participatory approaches.

Figure 6. In the dry-season boats have to be moved across the floodplain to remaining beels by other means, northwest Bangladesh.

Photo by Paul Thompson.

While there have been significant empowering changes in scattered locations for fishery and wetland management, participation in the water-management sector, particularly in larger polders, has been criticized for being imposed and regimented without actually empowering poorer floodplain households (Dewan, Buisson, & Mukherji, 2014).

Adaptation to Environmental and Climate Changes

Climate Change

Bangladesh is widely recognized as being on the front line of climate change (United Nations Development Programme, 2004) that is expected to exacerbate many of the natural hazards that already affect agriculture and fisheries. According to Ministry of Environment and Forests (2008, 2009), climate change is expected to result in

more frequent and severe tropical cyclones;

increased river bank erosion and sedimentation due to heavier, more erratic monsoon rainfall;

increased flooding because of greater snowmelt in upper catchments, higher monsoon rains, a rise in sea level, and storm surges;

dry-season crop stresses because of lower, more erratic dry-season rainfall, higher evapotranspiration, and increased salinity in the surface and ground water); and

saltwater intrusion into surface freshwater and aquifers and degradation of water quality.

These trends are predicted over a period of decades, to significantly reduce rice and wheat production in Bangladesh unless there are technological changes to adapt to these trends (Ministry of Environment and Forest, 2009). Moreover an increase in the middle of the predicted range of increase in average sea level of from 30 cm to 80 cm by 2100 (Ministry of Environment and Forest, 2009) would subject about 11% of the country to more regular inundation and adversely impact the lives and livelihoods of millions of people (estimates made in the early 2000s for 2100 are between 10 and 30 million), who it has been claimed could become environmental refugees. However, the extent of these changes has been challenged; for example, Brammer (2014) argues that the Bangladesh coastal floodplains are highly dynamic, with both sedimentation and land shrinkage and that the impact on livelihoods of any gradual relative sea level or salinity changes will be outweighed by other pressures, such as population.

Coastal Embankments

Coastal Bangladesh has a long history of vulnerability to hazards—cyclones, unusual high tides, floods, and saline water. The government response since the 1960s has been to build embankments and polders to protect agriculture land in a large area of low-lying lands in the coastal delta of the southwest, including areas inland of the Sundarbans mangrove forest. This enabled an expansion of rice farming, and then shrimp farming. Alarms have already been sounded about the fears that large-scale human displacement and mass migration from this area will result from increasing sea levels (Gore, 2009). However, physical processes and climatic trends are complex in this area and are less immediate threats compared with direct human pressures (Brammer, 2014). Channels outside of embankments continue to silt up as does the mangrove forest, keeping pace with sea level rise, whereas land within polders has shrunk (Auerbach et al., 2015). The result has been increasing drainage congestion, and after embankments were broken in cyclones, it left the polders tidally inundated for years, until the embankments were rebuilt. However, opening polders to tidal inflow due either to cyclones or community initiatives has been proven to cause a quite rapid rise of inside land levels from silt and to open up adjacent (outside) waterways.

Local communities proposed keeping tidal flows open, and in 1997 local people cut an embankment to reduce drainage congestion, against the Bangladesh Water Development Board’s wishes, but the measure was found to be effective. Moreover, it built on long-lost but traditional silt-management practices that had been used for smaller local embankments prior to the construction of large polders beginning in the 1960s. So retrospectively, a major project in this region incorporated this initiative in a concept that it formalized as “tidal river management” (Paul, Nath, & Abbas, 2013; Tutu, 2005). This involves in different years breaking the embankment of different polders to allow a tidal flow between the outside channels and that polder, at the cost of crop production inside that polder that year. This results in silt deposition inside the polder rather than in the outside channels. It was found to be socially acceptable and environmentally friendly solution, but one that requires a high level of coordination among many affected people. In any case, since the late 1990s, as noted, a complex hierarchy of water-management groups and associations has been established inside the polders with the aim of improving participatory water management, which also faces problems of elite influence and coordination (Bernier et al., 2016).

Adaptation and Integrated Floodplain Management

Government emphasis on building embankments has had negative impacts, despite having increased rice production. Local planning often creates interest in finding a more balanced approach that will address local needs for diverse floodplain products, as well as uncertainties and hazards. In response, a concept of integrated floodplain management (IFM) has been developed, tested, and promoted through action research with many CBOs in Bangladesh. IFM is intended to address challenges that affect floodplain livelihoods through the following initiatives:


Sluice-gate management—during the early dry season in October and November, sluice gates are opened to maximize quick drainage out of floodplain lands used for cultivation, leaving less water in the beels for fish.


Water abstraction for irrigation is the priority for rice farming in the dry season (January–May), so farmers pump water from beels and canals, but this reduces the critical dry-season habitat of fish and threatens floodplain aquatic life.


Early flash floods can damage dry-season rice crops before the harvest, so farmers keep sluice gates closed to protect crops, preventing migratory fish and their developing larvae from entering the floodplain.

The dry season is the critical time. Scarce water drives the productivity of a system in which water is overabundant in the wet season. Conflicts between farmers and fishers over the use of dry-season water are common. Wealthy, influential farmers tend to win such disputes at the expense of the poor and landless, whose livelihoods depend more on floodplain common-pool resources. Rich, influential farmers living close to structures tend to dominate decisions on opening and closing water-control infrastructure, and do this to support the crops and practices they traditionally use with little concern for impacts on poorer or more distant communities and no incentive to change crops. There are, however, alternative win-win approaches that can benefit farmers and poorer aquatic resource users.

Shankar et al. (2004) investigated the potential to maximize floodplain productivity and returns by taking a more integrated view of the resource base that would better balance agriculture with fisheries and benefit the poor. Modeling showed the potential to restore floodplain fisheries by modifying water management. Protecting and enhancing open capture fisheries most benefits the poor, but farmers also could gain by adopting profitable alternative crop-management practices that diversify risk and gain from subsistence fishing. This has been refined and brought into practice through work with CBOs.

Example of Operationalizing Integrated Floodplain Management

Linkages between elements of integrated floodplain management are shown in Figure 7. Within this evolving framework a range of actions have been tested and adopted in Bangladesh:

Cropping pattern management. Promoting dry-season crops that are less water demanding than irrigated rice to conserve water for fish and shorter-duration rice varieties to enable earlier sluice-gate opening to facilitate fish migration.

Modified sluice-gate management. Opening sluice gates in the pre- to early monsoon, when fish are migrating between rivers and floodplains for spawning (provided cropping patterns have changed to avoid crop losses).

Fishing effort control. Reducing fishing pressure, stopping fishing practices that target fish at critical times (e.g., barriers and lift nets close to sluice gates) or catch all fish (e.g., draining out waterbodies for fishing), establishing fish sanctuaries, and observing closed seasons for fishing.

Land retirement. Giving up high-risk crop cultivation in low land, leaving the land for fish.

Improving water quality and ecosystem health. Adopting practices of integrated pest management and enclosed jute processing to replace openwater processing which damages water quality (see Figure 8).

Habitat rehabilitation. Re-excavating waterbodies to hold water for overwintering fish, and re-establishing connections between floodplain beels and rivers.

Reintroduction of locally threatened species. Reintroduction and augmentation of low stocks of local rare and threatened fish species.

Watershed management. Implement of soil conservation measures and tree planting to reduce siltation and ensure that floodplain depressions will hold water.

Figure 7. Elements of integrated floodplain management (IFM).

Figure 8. Jute retting in open floodplain areas adversely affects water quality in Narail, southwest Bangladesh.

Photo by Paul Thompson.

Local institutions that can engage in collective action are a precondition for integrated floodplain management. In the largest piloting of IFM, Sultana and Thompson (2012) reported that among 280 existing CBOs, more than 300 farmers from 72 CBOs tried different low-water-demand alternative crops (garlic, potato, maize, sunflower, potato, wheat, and mustard) that were not previously grown in their areas, and found them to be profitable. They also reported that conservation measures (sanctuaries and closed seasons) had restored fish species diversity and catches, including 110 CBOs that established new or improved fish sanctuaries covering on average 6% of their dry-season water area, and that this was associated with improvements in food security. Overall, farmers and fishers in communities adopting IFM benefited from higher catches, higher incomes from crops, and greater community solidarity.

Adaptive Learning

Ideally, the many CBOs in Bangladesh floodplains would strive to continue improving their decisions and management of natural resources after project support ends, in response to environmental and economic variations and trends. Adaptive management and adaptive learning can be thought of as structured processes of “learning by doing” that emphasize systematic reflection on experiences and lessons, and testing management actions and interventions as a way of improving performance and reducing uncertainties. Adaptive co-management incorporates a hierarchy of institutional arrangements for sharing management responsibilities over scales of resource and an explicit commitment to learning among these partners (Armitage, Marschke, & Doubleday, 2008).

However, the isolation of each CBO dispersed across Bangladesh limited the scope of adaptive learning, which was limited to adjusting practices based only on its own experience. Networking between CBOs makes it possible to share lessons, accelerating adaptive learning. Adaptive learning networks have been proposed between individuals (Davidson-Hunt, 2006), or involved villages focused on technical aspects of resource management (Arthur & Garaway, 2005).

In 2007 an innovation was implemented in Bangladesh that has involved action research to test an organized learning network comprising over 250 existing CBOs that are managing floodplain natural resources (Sultana & Thompson, 2012). The process evolved through regular workshops and exchange visits. CBOs that had previously concentrated on either culture- or capture-fishery management or water management for rice collectively analyzed their experiences, including both the constraints each had encountered and the potential opportunities perceived, lessons learned, and good practices.

Typically, each CBO has an executive committee, which draws up a management plan and, in most years, updates it based on an informal review of its own experiences, sometimes bringing the general members into the discussion. The Bangladesh innovation was to establish an adaptive learning network to enhance this practice through the multiplier effect of sharing learning across CBOs. The CBOs could then share their experiences collectively, identifying constraints and knowledge gaps, coordinating piloting and changes in practice to address interlinked resource-management issues, and monitoring and assessing these changes.

Based on their experience facilitating this process, Sultana and Thompson (2012) concluded that face-to-face workshops or meetings of the CBO leaders from the same region were the most effective arrangement. The process that evolved is shown in Figure 9. Each CBO in a region sends a representative to two workshops a year, covering the cycle of activities shown in the bottom circle of figure 9. These workshops are used to identify common issues and uncertainties; solutions already proven by some CBOs; potential changes in their draft plans; and other activities that the CBOs want to improve or experiment with. The individual CBOs became more systematic in revising their plans based on their own experience and that of the other CBOs and their experiments. After attending workshops, CBO leaders share options with their members and changes to plans and actions are finalized by the executive committees of each individual CBO (top circle). This is coordinated in the network meetings so that alternative views can be tested in the form of comparative “experiments” where appropriate. In the workshops the CBOs also agree common indicators so that they can compare and assess impacts based on their own monitoring.

Figure 9. Concept for an adaptive learning network among community-based organizations.

Adapted from Sultana and Thompson (2012).

The outcomes were improvements in governance and natural-resource management. Most of the CBOs adopted win-win practices arising from a system-based view of the links between agriculture, water management, and fisheries—elements of integrated floodplain management—encouraged by sharing lessons. This has enhanced the overall productivity of floodplains and resilience to uncertain environmental conditions. Over the course of a four-year study, 84% of participating CBOs acted to improve fisheries management—mostly through measures such as creating fish sanctuaries and closed seasons that help sustain native wild fish populations and diversity. Moreover 62% of CBOs tested and promoted dry-season crops that were new to their areas (such as maize, sunflower, and garlic) and require a third or less of the water used by irrigated rice. This preserved more surface water in which fish could survive, enhanced the capacity to cope with drought, and yielded better financial returns than rice.

The network also encouraged innovation and improved linkages with service providers and, importantly, generated peer pressure among the CBOs to improve governance and natural-resource management. Improvements in governance included strengthening the internal management role of the disadvantaged within the CBOs, increasing the membership of women and their engagement in decision-making (Sultana & Thompson, 2008), further consulting with local stakeholders, and following transparent practices. This federation of CBOs develops capacity, confidence, and strength in numbers enabling communities to enter into dialogue with policy processes and attempt to resist threats to their access to common water resources.

Based on a series of assessments of the participating CBOs, in about 80% of locations where CBOs are active and members of the adaptive learning network, local people report that the CBOs have improved the livelihoods of poor people, and this is consistent across floodplain environments. More CBOs are seen as having protected or improved access for the poor since they became involved in adaptive learning, for example, allowing subsistence fishing by the poor, or ending exclusions imposed by previous leaseholders.



Research and development initiatives since the early 1990s in Bangladesh have revealed the complexity of interactions between ecosystems and livelihoods in the floodplain. The dominance of agricultural use of the floodplain, particularly to grow rice, and interventions to increase rice production have been challenged. Wild aquatic resources, particularly fish, and related ecosystem services have been shown to account for a substantial part of the economic value of floodplains, to contribute more to the livelihoods of the poor than to the livelihoods of larger landowners, and to be under increasing threat from intensification of both crop farming and aquaculture. Relevant to the international livelihood debates are findings as diverse as the nutritional significance of small wild fish as a source dietary micronutrients and the low fertility of silt deposition. Bangladesh has adopted and demonstrated successful methods to facilitate local participation in the planning and governance of floodplains. Empowering local communities and within them the poorer floodplain natural-resource users has been addressed by NGO initiatives, to some extent by policy decisions, and in action research. The Bangladesh experience highlights the role of collective action, provides lessons on building community-based organizations and then sustaining them, and reveals the scope to add value and build adaptation capacity through structured learning between communities.

Gaps and Research Questions

The links between rural floodplains and urban economic opportunities in the livelihood strategies of the poor are becoming increasingly apparent—for example, many floodplain households have members who periodically or seasonally migrate to urban centers or other rural districts to find work, or who leave to find work in times of crisis. The impact on floodplain livelihoods and the influence of such factors as climate stress remain topics for further research. The productivity and value of ecosystem services in a range of floodplain types and conditions (for example, considering flood depth, duration, and silt load) also deserves more research (research thus far has concentrated on only a few, less intensively farmed areas). Ultimately, the sustainability and resilience of floodplain ecosystems, including their natural resources and the rural livelihoods that depend on them remains in question.

In the face of climate change, the lessons and evidence generated in Bangladesh are directly relevant to environments and communities globally that face similar changes (such as salinity changes, flood risks, and unreliable rains). Of particular importance are the generic findings on governance and policies, local natural-resource governance, and the resilience of more diverse local agro-ecological systems and the livelihoods of the poor who depend on them.

Further Reading

  • Ali, M. Y. (1997). Fish, water and people. Dhaka, Bangladesh: University Press.
  • Brammer, H. (2010). After the Bangladesh Flood Action Plan: Looking to the future. Environmental Hazards, 9, 118–130.
  • Sultana P., & Thompson, P. (2012). Learning through networking: Enabling an adaptive learning network of local communities for integrated floodplain management in Bangladesh. In H. Ojha, A. Hall, & R. Sulaiman (Eds.), Adaptive collaborative approaches in natural resource governance: Rethinking participation, learning and innovation (pp. 138–176). London: Earthscan.
  • Thompson, P. M., Roos, N., Sultana, P., & Thilsted, S. H. (2002). Changing significance of inland fisheries for livelihoods and nutrition in Bangladesh. Journal of Crop Production, 6(1–2), 249–317.


  • Agrawal, A., & Ostrom, E. (2001). Collective action, property rights, and devolution of forest protected area management. In R. Meinzen-Dick, A. Knox, & M. Di Gregorio (Eds.), Collective action, property rights and devolution of natural resource management. Proceedings of the international conference held from 21–25 June, 1999 in Puerto Azul, Philippines (pp. 75–109). Feldafing: Deutsche Stiflung für internationale Entwicklung.
  • Akonda, A. W. (1989). Wetlands of Bangladesh. In D.A. Scott (Ed.), A directory of Asian wetlands (pp. 541–581). Gland, Switzerland: International Union for the Conservation of Nature, Worldwide Fund for Nature, International Council for Bird Preservation, International Waterfowl and Wetlands Research Bureau.
  • Ali, M. Y. (1990). Openwater fisheries and environmental changes. In A. A. Rahman, S. Huq, & G. R. Conway (Eds.), Environmental aspects of surface water systems of Bangladesh (pp. 145–165). Dhaka, Bangladesh: University Press.
  • Ali, M. Y. (1997). Fish, water and people. Dhaka, Bangladesh: University Press.
  • Armitage, D., Marschke, M., & Doubleday, N. (2008). Adaptive co-management and the paradox of learning. Global Environmental Change, 18(1), 86–98.
  • Arthur, R. I., & Garaway, C. J. (2005). Learning in action a case from small waterbody fisheries in Lao PDR. In J. Gonsalves, T. Becker, A. Braun, D. Campilan, H. de Chavex, E. Fajber et al. (Eds.), Participatory research and development for sustainable agriculture and natural resource management: A sourcebook (pp. 191–198). Laguna, Philippines: International Potato Center, and Ottawa, Canada: International Development Research Centre.
  • Atahar, S. A. (2013). Development project, land acquisition and resettlement in Bangladesh: A quest for well formulated national resettlement and rehabilitation policy. International Journal of Humanities and Social Science, 3(7), 306–319.
  • Auerbach, L. W., Goodbred, S. L., Mondal, D. R., Wilson, C. A., Ahmed, K. R., Roy, K. et al. (2015). Flood risk of natural and embanked landscapes on the Ganges-Brahmaputra tidal delta plain. Nature Climate Change, 5, 153–157.
  • Barr, J. J. F., Dixon, P.‑J., Rahman, M. M., Islam, M. A., Zuberi, M. I., McGlynn, A. A. et al. (2000). A participatory, systems-based, process for identification of improved natural resources management for better floodplain livelihoods (Project Report of R6756). Centre for Land Use and Water Resources Research, University of Newcastle, U.K.
  • Benson, C., & Clay, E. J. (2002). Bangladesh: Disasters and public finance (Disaster Risk Management Working Paper Series 5). World Bank, Washington, DC.
  • Berkes, F., & Farvar, M. F. (1989). Introduction and overview. In F. Berkes (Ed.), Common property resources: Ecology and community-based sustainable development (pp. 1–17). London: Bellhaven Press.
  • Berkes, F., George, P., & Preston, R. (1991). Co-management: The evolution of the theory and practice of joint administration of living resources. Alternatives, 18(2), 12–18.
  • Bernier, Q., Sultana, P., Bell, A. R., & Ringler, C. (2016). Water management and livelihood choices in southwestern Bangladesh. Journal of Rural Studies, 456, 134–145.
  • Borrini-Feyerabend, G., Farvar, M. T., Nguinguiri, J. C., & Ndangang, V. A. (2000). Comanagement of Natural Resources: Organising, Negotiating and Learning-by-Doing. GTZ and IUCN, Kasparek Verlag, Heidelberg (Germany). Reprint 2007 [first publication in 2000]. GTZ STANDS FOR Deutsche Gesellschaft für Technische Zusammenarbeit.
  • Boyce, J. K. (1990). Birth of a megaproject: Political economy of flood control in Bangladesh. Environmental Management, 14, 419–428.
  • Brammer, H. (1990). Floods in Bangladesh: Geographical background to the 1987 and 1988 floods. Geographic Journal, 156(1), 12–22.
  • Brammer, H. (2000). Controversies surrounding the Bangladesh Flood Action Plan. In D. J. Parker (Ed.), Floods (Vol. 1, pp. 302–315). London: Routledge.
  • Brammer, H. (2004). Can Bangladesh be protected from floods? Dhaka, Bangladesh: University Press.
  • Brammer, H. (2010). After the Bangladesh Flood Action Plan: Looking to the future. Environmental Hazards, 9, 118–130.
  • Brammer, H. (2014). Bangladesh's dynamic coastal regions and sea-level rise. Climate Risk Management, 1, 51–62.
  • Cooke, B., & Kothari, U. (2002). The case for participation as tyranny. In B. Cooke & U. Kothari (Eds.), Participation: The New Tyranny? (pp. 1–15). London: Zed Books.
  • Davidson-Hunt, I. J. (2006). Adaptive learning networks: Developing resource management knowledge through social learning forums. Human Ecology, 34, 593–614.
  • Deba, A. K., Haque, C. E., & Thompson, S. (2014). Man can’t give birth, woman can’t fish: Gender dynamics in the small-scale fisheries of Bangladesh. Gender, Place and Culture, 22(3), 305–324.
  • Dewan, C., Buisson, M.‑C., & Mukherji, A. (2014). The imposition of participation? The case of participatory water management in coastal Bangladesh. Water Alternatives, 7, 342–366.
  • Environment and GIS Support Project for Water Sector Planning. (1997). Floodplain fish habitat study. Dhaka, Bangladesh: Water Resources Planning Organization.
  • Environment and GIS Support Project for Water Sector Planning. (2000). Riverine Chars in Bangladesh: Environmental dynamics and management issues. Dhaka, Bangladesh: University Press.
  • Elahi, K., Ahmed, K. S., & Mafizuddin, M. (1991). Riverbank erosion, flood and population displacement in Bangladesh. Jahangirnagar University Riverbank Erosion Impact Study, Dhaka, Bangladesh.
  • FAP 16/19. (1993). Charland Study Overview: Summary Report. Dhaka, Bangladesh: Flood Plan Coordination Organization, Ministry of Irrigation Water Development and Flood Control, Dhaka, prepared by Irrigation Support Project for Asia and the Near East.
  • Flood Hazard Research Center Bangladesh. (2013). Risk coping and collective action in Bangladesh Floodplains (Briefing Paper no. 12). Dhaka, Bangladesh.
  • Fourth Fisheries Project. (2005). Key lessons and learning from inland open-water fisheries. Report no 4. Fourth Fisheries Project, Department of Fisheries, Dhaka, Bangladesh.
  • Gore, A. (2009). An inconvenient truth. London: Bloomsbury.
  • de Graaf, G., Born, B., Uddin, A. M. U., & Marttin, F. (2001). Floods, Fish and Fishermen. Dhaka, Bangladesh: University Press.
  • Halls, A. S. (1998). Impact of flood control schemes on river fish migrations and assemblages in Bangladesh. Journal of Fish Biology, 53, 358–580.
  • Hanchett, S. (1997). Participation and policy development: The case of the Bangladesh Flood Action Plan. Development Policy Review, 15, 277–295.
  • Hardin, G. (1968). The tragedy of the commons. Science, 162, 1243–1248.
  • Hillel, D. (1992). Out of the earth: Civilisation and the life of the soil. London: Aurum.
  • Holmes, T., & Scoones, I. (2000). Participatory policy processes: Experiences from north and south (IDS Working Paper 113). Institute of Development Studies, Brighton.
  • Hossain, A. N. H. A. (2006). The impact of floods on Bangladesh and options for mitigation. In K. U. Siddiqui, & A. N. H. A. Hossain (Eds.), Options for flood risk and damage reduction in Bangladesh (pp. 55–70). Dhaka, Bangladesh: University Press.
  • Hossain, M. (2015). Improving land administration and management in Bangladesh. Paper prepared for the General Economics Division of the Planning Commission. Dhaka, Bangladesh: Bangladesh Institute of Development Studies.
  • Hughes, R., Adnan, S., & Dalal-Clayton, B. (1994). Floodplains or flood plans? London: International Institute for Environment and Development and Research Advisory Services.
  • International Union for the Conservation of Nature Bangladesh. (2000). Red book of threatened fishes of Bangladesh. Dhaka, Bangladesh: IUCN—the World Conservation Union.
  • International Union for the Conservation of Nature Bangladesh. (2006). Natural resource economic evaluation of Hakaluki Haor. Report submitted to Ministry of Environment and Forests BGD/99/G31- Coastal and Wetland Biodiversity Management Project: Hakaluki Haor Component, Dhaka, Bangladesh.
  • International Union for the Conservation of Nature Bangladesh. (2015). Red list of Bangladesh volume 5: freshwater fishes. Dhaka, Bangladesh: IUCN—the World Conservation Union.
  • Mehta, L., Leach, M., Newell, P., Scoones, I., Sivaramakrishnan, K., & Way, S. (1999). Exploring understandings of institutions and uncertainty: New directions in natural resource management (IDS Discussion Paper 372). University of Sussex, Brighton.
  • Meinzen-Dick, R. S., & Di Gregorio, M. (Eds.). (2004). Collective action and property rights for sustainable development. (2020 Focus 11). Washington DC: International Food Policy Research Institute.
  • Miah, M. M. (1988). Flood in Bangladesh. Dhaka, Bangladesh: Academic.
  • Ministry of Environment and Forest. (2008). Bangladesh climate change strategy and action plan. Dhaka, Bangladesh: Government of the People’s Republic of Bangladesh.
  • Ministry of Environment and Forest. (2009). National Adaptation Program of Action (NAPA). Dhaka, Bangladesh: Government of the People’s Republic of Bangladesh.
  • Ministry of Water Resources. (1999). National water policy. Dhaka, Bangladesh: Government of People’s Republic of Bangladesh.
  • Ministry of Water Resources. (2001). Guidelines for participatory water management. Dhaka, Bangladesh: Government of the People’s Republic of Bangladesh.
  • Minkin, S. F., Rahman, M. M., & Halder, S. (1997). Fish biodiversity, human nutrition and environmental restoration in Bangladesh. In C. Tsai & M. Y. Ali (Eds.), Openwater Fisheries of Bangladesh (pp. 183–198). Dhaka, Bangladesh: University Press.
  • Mosse, D. (2002). People’s knowledge, participation and patronage: Operations and representations in rural development. In B. Cooke & U. Kothari (Eds.), Participation: The new tyranny? (pp. 16–35). London: Zed Books.
  • Muir, J. (Ed.). (2003). Fisheries sector review and future development: Theme study; economic performance. Dhaka, Bangladesh: World Bank, Danida, USAID, FAO, and DFID.
  • Mustafa, M. G., & Brooks, A. C. (2009). A comparative study of two seasonal floodplain aquaculture systems in Bangladesh. Water Policy, 11(Suppl. 1), 69–79.
  • Ostrom, E. (1990). Governing the commons: The evolution of institutions for collective action. Cambridge, U.K.: Cambridge University Press.
  • Ostrom, E., & Schlager, E. (1996). The formation of property rights. In S. Hanna, C. Folke, & K. Mäler, (Eds.), Rights to nature (pp. 127–156). Washington, DC: Island Press.
  • Paul, A., Nath, B., & Abbas, M. R. (2013). Tidal river management (TRM) and its implication in disaster management: A geospatial study on Hari-Teka river basin, Jessore, Bangladesh. International Journal of Geomatics and Geosciences, 4(1), 125–135.
  • Pinkerton, E. (Ed.). (1989). Co-operative management of local fisheries. Vancouver, Canada: University of British Colombia Press.
  • Rahman, A. K. A. (2005). Freshwater fish of Bangladesh. Dhaka, Bangladesh: Dhaka University.
  • Rahman, M., Capistrano, D. A., Minkin, S. F., Islam, A., & Halder, S. (1999). Experience of community managed wetland habitat restoration. In H. A. J. Middendorp, P. M. Thompson, & R. S. Pomeroy (Eds.), Sustainable inland fisheries management in Bangladesh (pp. 111–121). ICLARM Conf. Proc. 58. Manila: ICLARM.
  • Ramsar. (2001). Wetland values and functions. Gland, Switzerland: Ramsar Convention Secretariat.
  • Rashid, H., & Pramanik, M. A. H. (1993). Areal extent of the 1988 flood in Bangladesh: How much did the satellite imagery show? Natural Hazards, 8, 189–200.
  • Report of the Task Forces on Bangladesh Development Strategies for the 1990s. (1991). Dhaka: University Press Ltd.
  • Shankar, B., Halls, A., & Barr, J. (2004). Rice versus fish revisited: On the integrated management of floodplain resources in Bangladesh. Natural Resources Forum, 28, 91–101.
  • Soussan, J., & Datta, A. (1998). Community partnership for sustainable water management: Experiences of the BWDB Systems Rehabilitation Project. Vol 1, summary report. Dhaka, Bangladesh: University Press.
  • Sultana, P. (2012). Implications of floodplain aquaculture enclosure. Journal of Environmental Planning and Management, 55, 1159–1174.
  • Sultana, P., & Abeyasekera, S. (2008). Effectiveness of participatory planning for community management of fisheries in Bangladesh. Journal of Environmental Management, 86, 201–213.
  • Sultana, P., Johnson, C., & Thompson, P. (2008). The impact of major floods on flood policy evolution: insights from Bangladesh. International Journal of River Basin Management, 6(special issue), 1–10.
  • Sultana, P., & Thompson, P. M. (1997). Impact of flood control and drainage on fisheries in Bangladesh and the design of mitigating measures. Regulated Rivers, 13, 43–55.
  • Sultana, P., & Thompson, P. M. (2004). Methods of consensus building for community-based fisheries management in Bangladesh and the Mekong Delta. Agricultural Systems, 82, 327–353.
  • Sultana, P., & Thompson, P. M. (2008). Gender and local floodplain management institutions: A case study from Bangladesh. Journal of International Development, 20, 53–68.
  • Sultana, P., & Thompson, P. M. (2010). Local institutions for floodplain management in Bangladesh and the influence of the Flood Action Plan. Environmental Hazards, 9(1), 26–42.
  • Sultana P., & Thompson P. M. (2012). Learning through networking: Enabling an adaptive learning network of local communities for integrated floodplain management in Bangladesh. In H. Ojha, A. Hall, & R. Sulaiman (Eds.), Adaptive collaborative approaches in natural resource governance: Rethinking participation, learning and innovation (pp. 138–176). London: Earthscan.
  • Thilsted, S. H., Roos, N., & Hasan, N. (1997). The role of small indigenous fish species in food and nutrition security in Bangladesh. NAGA: The ICLARM Quarterly, 20(3–4), 82–84.
  • Thompson, P., & Colavito, L. (2007). Economic value of Bangladesh wetlands. (MACH Technical Paper 6). Dhaka, Bangladesh: Winrock International.
  • Thompson, P. M., Roos, N., Sultana, P., & Thilsted, S. H. (2002). Changing significance of inland fisheries for livelihoods and nutrition in Bangladesh. Journal of Crop Production, 6(1–2), 249–317.
  • Thompson, P. M., & Sultana, P. (2000). Shortcomings of flood embankment strategies in Bangladesh. In D. J. Parker (Ed.), Flood (Vol. 1, pp. 316–333). London: Routledge.
  • Thompson, P. M., Sultana, P., & Islam, N. (2003). Lessons from community-based management of floodplain fisheries in Bangladesh. Journal of Environmental Management, 69(3), 307–321.
  • Toufique, K. A., & Gregory, R. (2008). Common waters and private lands: Distributional impacts of floodplain aquaculture in Bangladesh. Food Policy, 33, 587–594.
  • Tutu, A. U.‑A. (2005). River management in Bangladesh: A people’s initiative to solve water-logging. Participatory learning and action, 51, 117–123.
  • United Nations Developmental Programme. (2004). A global report: Reducing disaster risk: A challenge for development. Retrieved from
  • United Nations Conference on Environment and Development (1992). The Rio declaration on Environment and Deveopment (1992).
  • World Bank. (1990). Flood control in Bangladesh: A plan for action. Washington, DC: Asia Region Technical Department.
  • World Bank. (1991). Bangladesh fisheries sector review (Report No. 8830-BD). Washington DC: World Bank.
  • WorldFish Center (2003). Community-based fisheries management Phase 2 (CBFM-2) Annual Report September 2001–December 2002. Dhaka, Bangladesh: WorldFish Center, Department of Fisheries, Banchte Sheka, BELA, BRAC, Caritas, CNRS, CRED, FemCom, Proshika.