World Programme - hydro-social-deltas.un-ihe.org...(Understanding flows of water and people to improve policies and strategies for disaster risk reduction ... 1.2 Flood Risks and 2016

Urbanising Deltas of the World Programme of NWO

Hydro‐Social Deltas Project (Understanding flows of water and people to improve policies and strategies for disaster risk reduction and sustainable development of delta areas in the Netherlands and Bangladesh)

Migration and flood‐erosion hazards: influence of local institutions and infrastructure

August 2017 Parvin Sultana and Paul Thompson Flood Hazard Research Centre Bangladesh House 97 Road 6A Old DOHS Banani Dhaka Bangladesh <[email protected]

Hydro-Social Deltas ii Migration and hazards 2016

CONTENTS

Page Executive Summary iv 1 Introduction 1-1 1.1 Study Aim 1-1 1.2 Flood Risks and 2016 Floods in Context 1-1 1.3 Evacuation and Migration 1-3 1.4 Riverine Chars 1-3 2 Methods 2-1 2.1 Focus Group Discussions 2-1 2.2 Sample Household Survey and Characteristics of Households 2-1 3 Lalmonirhat (mostly mainland) 3-1 3.1 Context 3-1 3.2 2016 floods 3-3 3.3 2016 Flood and Erosion Impacts on Households 3-4 3.4 Role of Embankments 3-5 4 Gaibandha (mainly riverside and chars) 4-1 4.1 Context 4-1 4.2 2016 floods 4-3 4.3 Losses from 2016 Floods and Erosion 4-4 4.4 Role of Embankments 4-5 5 Kurigram (mainly chars separated from mainland) 5-1 5.1 Context 5-1 5.2 2016 floods 5-3 5.3 Flood and Erosion Losses in 2016 5-5 5.4 Role of Embankments 5-5 6 Responses in 2016 Floods 6-1 6.1 Household Responses and Coping with Floods and Erosion 6-1 6.2 Assistance Received 6-5 7 Migration and Movements 7-1 7.1 Home Location 7-1 7.2 Incidence of Different Types of Movement 7-2 7.3 Short-term Hazard Response Movement in 2016 (Evacuation) 7-2 7.4 Long-term Hazard Response Movement 7-5 7.5 Livelihood Related Movement 7-6 7.6 Role of Communication in Seasonal Movement 7-15 8 Trends and Changes over Time 8-1 8.1 Lifetime Movements 8-1 8.2 Hazards in Past Five Years 8-1 8.3 Changes in Incomes 8-3 8.4 Livelihood Changes 8-5

Hydro-Social Deltas iii Migration and hazards 2016

Page 9 Overall Findings and Conclusions 9-1 9.1 Significance and Impact of 2016 Floods 9-1 9.2 Role of Embankments 9-1 9.3 Role of Local Institutions 9-2 9.4 Coping Strategies 9-2 9.5 Role of Migration 9-2 9.6 Implications of Findings 9-3 References R1 Annex 1

Detailed Records of Focus Group Discussions A1-1

Annex 2

Survey Formats A2-1

ACKNOWLEDGEMENTS With thanks to Habib Ahmed, the concerned CBO leaders, and our field team for assistance in focus group discussions and interviews; and to all of the flood and erosion affected people who were interviewed or who participated in discussions in the study areas for giving their time and sharing their experiences.

ABBREVIATIONS CBO Community Based Organisation FAP Flood Action Plan FGD Focus Group Discussion FHRC Flood Hazard Research Centre NGO Non-Governmental Organisation

Hydro-Social Deltas iv Migration and hazards 2016

EXECUTIVE SUMMARY

This study aimed to fill knowledge gaps regarding how households use temporary evacuation, seasonal and more permanent migration as elements of coping strategies; the role of local community organisations in strengthening coping; and the influence on coping of embankments. Too do this we investigated the impacts of floods and erosion in 2016 and more generally on communities and households living in the Brahmaputra River and Teesta River floodplains. Focus group discussions were held in three communities, each with existing community based organisations in each of three districts (Lalmonirhat, Kurigram and Gaibandha). In one representative site in each district from among these a sample of households were interviewed (chars or islands within the braided river in Kurigram, settlements outside and on the embankment in Gaibandha, and protected by embankments in Lalmonirhat). All of the respondents have experienced floods and erosion in the past – having moved home typically 2-10 times in their lifetime, even those living currently in locations at low erosion risk proved to have moved there due to past river erosion elsewhere. Over the past five years on average households report that their livelihoods have improved based on rising incomes from sources in their villages and from migration (more days worked and increasing wages), but landholdings have declined due to erosion, and costs also increase, and hazard events caused loss of assets, so that most households report running down savings. Hence while immediate coping ability appears to be improving, long term resilience may be declining. In all areas the 2016 floods were more severe than recent past years and involved a series of flood peaks. In Lalmonirhat a smaller percentage of households were flooded than in the other sites, but affected households evacuated to embankments for about two weeks. In Gaibandha most of the households were flooded and the majority of those moved for up to a month to embankments and protected areas, including households affected by embankment breaches. In Kurigram although most households were flooded responses varied between locations – some evacuated to embankments particularly those affected by erosion, but many people sheltered on and by their homes. Immediate coping involved eating less and poorer quality food for almost all households. Many households obtained cash to survive the effects of loss of property and crops, moving, and loss of work by borrowing, selling livestock and/or migrating for work. Key elements of coping strategies include using embankments as shelter, keeping livestock which can be sold, and social capital in the form of trusted relatives and neighbours. The latter has been strengthened by community organisations which provided information and warnings, helped people move homes and livestock, and helped in accessing services through local government and NGOs. Government, and to a lesser extent NGOs, provided important assistance particularly in the form of food aid, but was not reported to have performed well in maintaining law and order. Seasonal migration for work is reported to have increased from these areas. Over half of households are involved with men moving for half or more of the year to pull rickshaws or work as labourers in urban centres (mainly greater Dhaka but also towns in the northwest), or going to districts that face shortages of farm labourers in peak seasons (notably from Kurigram to Munshiganj). Migrants often move in groups and have built up personal contacts in destinations complemented by use of mobile phones. The lowest incidence of such migration was in July-August (when men came back to their villages to help their families cope in the flood season), and was highest in October as part of recovery after floods. But migrants commonly visit home at approximately monthly intervals. While such migration

Hydro-Social Deltas v Migration and hazards 2016

appears to improve coping ability, it involves travel costs, leaving wives and families, and living in poor conditions notably lacking in sanitation in their destinations. The study shows how households living in riverine areas make use of complex hydro-social landscapes. Physically making use of embankments as temporary flood shelter and more long term as shelter close to their lost land after erosion, and making use of work opportunities in expanding urban areas and other districts. Socially cooperating with others from their community to cope locally, and to find work, as well as involving to some extent in community initiatives for local coping. This reveals implications and opportunities for government and NGOs – to design embankments that are more effective as flood shelters, to improve sanitation and housing areas for seasonal migrants in urban areas, to help support community organisations, and to strengthen and improve women’s lives considering that many men are absent from these areas for much of the year.

Hydro-Social Deltas Migration and hazards 2016 1-1

1 INTRODUCTION 1.1 Study Aim The aim of this study was to investigate the impacts of floods and erosion in 2016 in particular, and more generally, on communities and households living in the active Brahmaputra floodplain. The study focused on adding to existing knowledge and filling knowledge gaps regarding how households use temporary evacuation, seasonal and more permanent migration/relocation as elements of coping strategies; the role of community initiatives and local community organisations in strengthening coping; and the influence on coping of flood protection infrastructure (embankments). The hydro-social deltas project aims to advance understanding of the complex web of interactions between dynamic physical and social process in the Bangladesh floodplain-delta system including the links and feedbacks between floods, urbanization and government responses. This research contributes to the hydro-social deltas project and framework by investigating flood and erosion coping with a focus on how people use migration between rural flood prone areas and urban economy, and the role of structural flood mitigation responses (embankments) in coping comparing riverine unprotected and embankment protected hydro-social landscapes already identified in the project. 1.2 Flood Risks and 2016 Floods in Context Bangladesh experiences flooding on up to two-thirds of its territory (Brammer, 2004). During the monsoon months (June to September), when 80% of annual rainfall occurs, the Ganges, Brahmaputra and Meghna rivers bring flood waters and about a billion tonnes of sediment per year into Bangladesh from an upstream catchment area 12 times the area of the country. Rainfall within Bangladesh accounts for a further large element of potential flood waters. It is important to differentiate between the Bangali concepts: ‘bonna’ (i.e. floods) and ‘barsha’ (i.e. normal monsoon conditions including inundation). Rural life is well adapted to the normal monsoon patterns of inundation, known as “barsha”, but if the frequency, timing, intensity, or duration of inundation is atypical and exceeds 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. Some normal annual flooding is important to the people and economy of Bangladesh. To a large extent, communities have adapted and developed resilience to the expected variations in the normal annual water cycle, including the expected timing and range of water levels. Cropping patterns, choices of crops and timing of planting and harvesting are all adjusted to normal floods, which also replenish groundwater, provide fertile growing conditions, and support important capture fisheries. From the breakdown of Bangladeshi land types (Table 1.1), it is evident that all land types except high lands are subjected to inundation to different degrees. Normally, 20–25% of the country is inundated during every monsoon season. However, severe floods of high magnitude can have adverse impacts on the economy and can cause major loss of life. In the case of extreme flood events, 40–70% of the country’s area can be inundated, as shown in the extreme flood events of 1954–55, 1974, 1987–88, and 1998. With increases in population, rapid urbanisation, growth of infrastructure, other economic development, and poor maintenance of flood mitigation infrastructure, vulnerability to floods has been increasing. There has also been extensive debate since the 1990s over the role of embankments in enabling economic growth and development (Boyce, 1990;


Brammer, 2000), their environmental impacts particularly on fisheries (Ali, 1997; de Graaf et al., 2001), and the more general scope for living with floods with or without structural mitigation works (Hughes et al., 1994; Sultana et al, 2008; Thompson and Sultana 2000).

Table 1.1 Land types in Bangladesh in relation to typical flood depths (from FAO, 1988)

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 estimated to have been 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 were damaged, and losses were estimated at up to US$2.8 billion (Brammer, 2004; Hossain, 2006). It has been argued that catastrophic floods can have major adverse consequences for long-term development of the country (Brammer, 1999), although 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) has improved. This suggests that economic resilience has improved at a macro-economic level. Although floods in 2016 did not attract the level of media coverage accorded to floods in previous decades, Bangladesh suffered very severe local monsoon floods during July to September along the Brahmaputra-upper Jamuna River and its tributaries (Teesta, Dharala, Dudkumar). Intense heavy rainfall occurred in the foothills of the Himalayas in India from Uttarkhand to Arunachal and Meghalaya during the second week of July, generating high runoff into the rivers flowing into northwest Bangladesh. The 2016 flood in the upper reaches within Bangladesh of the Brahamputra-Jamuna and its tributaries has been estimated to be have about a 1-in-100 year return period as the water level at the Bahadurabad gauging station (located in this area) crossed the previous highest ever recorded water level (from 1988). As will be shown, this study found that flood affected people experienced more than one flood peak. However, overall the durations of 2016 flood peaks were less than the extended high flood experienced in 1998 flood, for several reasons. The water level of the River Ganges was much below the danger level, so there was no backing up effect on the Jamuna River, enabling flood water to flow faster into the Bay of Bengal. Also the main flood peak coincided with low neap tide conditions in the Bay of Bengal creating a steeper gradient for the flood water. Future flood risks in Bangladesh are likely to increase. Because of global warming, the world temperature is continuously rising. For example, 2015 was the warmest year since 1880, and 2016 then surpassed that record. Higher temperatures on average increase water


holding capacity in the atmosphere, increasing the probability of heavy rainfall events in the future, which is one factor likely to increase the incidence of unusual floods. Similarly increasing temperatures mean that Himalayan snowmelt is increasing which is expected to bring earlier and higher river flows. Erosion is common along the banks of the main rivers and into the coastal zone (Haque and Zaman, 1989). The Jamuna–Brahmaputra river has been widening and shifting its course since 1830, becoming braided. This trend has continued during the period for which satellite images are available, with the monsoon extent of the river widening by about 128 m per year during 1973–2000 (EGIS, 2000). This means that lands along the river banks that have been settled for generations continue to be eroded, and also the islands (chars) within the braided channel are often subject to erosion. In general bank erosion occurs during the monsoon season, during high floods and also when water levels are rising or falling in the pre- and post-monsoon periods. 1.3 Evacuation and Migration Hazard research in the past has treated evacuation as a result of a positive decision to respond to the impending and immediate threat. The key variables influencing a decision to evacuate – in many different cultural circumstances - include the presence of a warning system, the credibility of the warning message, the perceived personal risk presented by the hazard and the logistics of the evacuation itself (Perry, 1979). Demographics are also seen as important, including gender, age, household income, educational attainment and ethnicity (Paul et al., 2010; Elliott and Pais, 2006). The evacuation decision is also theorised as a complex social process, involving family, community and other group interests and interactions (see Falk et al., 2006). It is expected that a similar but expanded set of factors will underlie the more complex decisions in Bangladesh regarding temporary evacuation during flood events, seasonal and temporary movements for work as part of coping strategies, and longer-term movements forced by erosion and permanent loss of land. Although evacuation in response to cyclones and coastal flooding have been well studied in Bangladesh (reviewed for example in Penning-Rowsell et al., 2013), riverine floods and erosion lead to different responses due to the nature of the losses and impacts, the physical features of the area affected, and its settlement history. For example, over 80% of char households (see next section) in the Brahmaputra–Jamuna river reported building raised platforms (macha) inside their homes in the major flood of 1988 (Thompson and Tod, 1998). Nevertheless, in severe (bonna) floods there may be no safe shelter on the river chars – about one-quarter of households in some char villages took shelter on boats in 1988 (FAP 14, 1992). Erosion is a long-term or permanent loss of land along with any immovable property on that land; as described by Haque and Zaman (1989) in this case ‘staying put’ is not an option. When both homestead and land are eroded/ submerged for long periods on the chars, the affected people may move as a whole village (retaining the same village name) to a new location, even some distance away (FAP 16, 1995). 1.4 Riverine Chars The islands within the braided channels of the main rivers of Bangladesh and towards the estuarine end of the delta system are known as chars in Bengali: the “mid-channel island that periodically emerges from the riverbed as a result of accretion” (Elahi et al. 1991). Chars have been distinguished from other alluvial lands, although the process of formation is similar, for example:


By Diara, generally means, the low bank of a river. Here it is used for any alluvial accretion on the banks of any waterbody. Char, usually means, any accretion in a river. Here it is used only for islets in the rivers (Rashid 1991, p 17).

However, the term diara has not received wider usage in flood and erosion coping studies in Bangladesh. Studies under the Flood Action Plan in the early 1990s (FAP 3.1 1992; FAP 16/19 1993) redefined, for planning purposes, charland. They defined charland as including not only islands (“island chars”) and accreted deposits along the river banks (“attached chars”), but also the strips of mainland along the river banks that are prone to flooding and erosion and are not protected by embankments or similar infrastructure (such as roads). This same classification based on local terminology and knowledge in the Padma chars was adopted by Baquee (1998). Fig. 1.1 illustrates this definition.

Figure 1.1 Charland Classification

The chars are the most flood prone environment in Bangladesh, and this has shaped livelihoods. In extreme events by definition the entire riverine charlands are inundated, since this area is located within the active river channels. Flood protection is not feasible in this highly dynamic environment where land erodes and also accretes and river channels within the active floodplain are prone to shifting. Although life in the charlands is dominated by floods and the seasonal rise and fall of river levels, when charland is above river level and thus potentially cultivable, the sandy soils that cover significant areas dry out and are only suited to crops such as millets and groundnuts. In the Brahmaputra-Jamuna the dominant crops traditionally were single cropping of boro (dry season) rice, a single millet crop, and mixed broadcast aus (early monsoon) and aman (late monsoon) paddy. A notable feature of accreted charland is the extent of grasses known as khaisa or catkin grasses (Saccharium sp.). These grasses are used as grazing/cut-and-carry fodder, for fuel, and as one of the main house construction materials in the chars – both as thatch and (using the older stems) for walls. Grasses form the initial stage in the sequence of converting new chars to cultivable soil, and are the natural colonists of chars. Livestock are a key asset for char people, they not only offer a means of adding value to the grasses that colonise recently accreted land before cultivation is possible, but they are also a mobile asset that can be moved to safe locations when land is eroded or flooded. Nevertheless livestock are


also vulnerable to floods, theft, and disease, and these problems are more significant in char areas than mainland because of the lack of services to counter these threats. Fishing is an important but not dominant economic activity on the charlands. In the study area, as in other parts of Bangladesh, since the late 1990s to early 2000s a number of community based organisations (CBOs) have been established under NGO and government programmes. Prior to this period NGOs had been reluctant to work in these areas because their model of group based savings and loans underpinned by group-village collateral (savings and social cohesion and norms) was not effective because erosion results in households dispersing from former settlements, or even moving together to a new location. However, recognizing the high vulnerability and needs of these areas government projects and associated NGOs and NGO projects attempted to reduce vulnerability through programmes such as asset transfer to poorer households (e.g. providing cattle), raising the plinths of houses, improving access to services such as education and health. RDRS (originally Rangpur Dinajpur Rural Services) developing saving and revolving fund systems over larger areas roughly coinciding with the jurisdiction of Union Parishads (local councils) by facilitating community based organisations (CBOs) known as “federations” serving several hundred households that may move location in response to floods and erosion within that area but usually maintain a presence there. Other initiatives focused more on local natural resource management such as CBOs formed to access and manage capture fisheries in the river. As part of their services to members, some of these CBOs reported that they have taken an active role in helping members cope when their homes are affected by floods and erosion, and the study investigated this. Much of the existing literature on char issues is from the 1990s (Baqee 1998; Elahi et al. 1991; FAP 16/19 1993), when urbanization and inter-district communications in Bangladesh were less developed than at present, and CBOs did not exist in these areas. This study adds to knowledge and focuses in part on urban-rural linkages and their role in coping with floods and erosion. It is based on case studies of communities affected by flood and erosion to the west of the Brahmaputra-Jamuna River in Bangladesh through fieldwork conducted within a few months of the 2016 flood event.


2 METHODS The study is based on two methods employed in two stages - firstly focus group discussions in each of nine communities /CBO areas (see Annex 1), and secondly interviews with a sample of households. 2.1 Focus Group Discussions The focus group discussions were conducted in three districts (Fig. 2.1), with three communities in each district, and were intended to update existing information and fill gaps in knowledge (on hazard experience, trends, and responses of local people and CBOs) as general context. They then focused on investigating in greater depth events in 2016, specifically:

the main migration patterns (if any) adopted by households (disaster response, short term, seasonal, long term);

migration responses to this year's floods and erosion (e.g. which households engage in migration and which do not migrate, and the reasons);

the relative advantages and disadvantages of migration; the vulnerability of locations people migrate to; the influence of flood protection infrastructure on these communities (whether outside

or behind it, and whether this is used for temporary shelter, affected flood exposure, traps water, is vulnerable to breaches, etc);

the extent that CBOs have responded to this year's flooding and erosion, and whether this has helped households (and which ones); and

any change in resilience or household strategies. The focus groups were conducted in mid October in nine locations noted in the following sections. In consultation with the community based organisations active in each area, and with which FHRC has worked for several years, a suitable location was selected in advance in or near one of the villages most affected by floods and erosion in 2016. In each location the study leader and research associate met with local people to discuss their experience following a pre-defined checklist. The groups were often large (30 to 40 persons) and more than an ideal number for an interactive discussion, but it was not possible to turn people away as flood and erosion affected people were often living crowded together, people were interested to recount their experiences, and also in some cases people came hoping to receive relief. 2.2 Sample Household Survey and Characteristics of Households Using the information obtained in the focus groups, the team purposively selected one area from among the three covered in each district that had been more severely affected by floods in 2016 and represented one of the three categories of socio-hydrological landscapes identified: chars, lands on the riverside of embankments (including those where people were sheltering on embankments), and areas protected by embankments. In each of these three areas 40 households were sampled at random from the households identified in that area. The senior member of the household was interviewed using a pre-tested questionnaire (see Annex 2). The aim of the household survey was to generate more quantitative evidence on the impacts and responses to floods in 2016 including but not limited to migration, as well as information on the past experience of floods and erosion of households.


Fig. 2.1 Map of Bangladesh showing three districts where the study was conducted

The breakdown of sample households by ownership of a homestead and its location is shown in Table 2.1. The sample household locations are consistent with the main socio-hydrological landscapes represented in that district and investigated in the FGDs, except that in Gaibandha in the sample survey relatively more households were interviewed who

Kurigram

Lalmonirhat

Gaibandhat


are now living on the embankment having been displaced from erosion prone areas on the riverside. Table 2.1 Location of own homestead land of sample households (number of households)

District Char Between river and embankment

Behind embankment

Other Total

Lalmonirhat 3 0 29 8 do not have their own homestead

but live behind bank 40

Gaibandha 0 18 0 22 live on embankment 40

Kurigram 34 0 0 6 live on other persons’ land in

chars 40

Most of the respondents were male household heads but just over a quarter of the respondents were women - either household heads or de facto responsible as men from those households were absent working elsewhere. Most respondents were the age range 31 to 50 years (Table 2.2).Over half of all respondents in each of the three areas are illiterate, but secondary education appears to be more frequent in the Gaibandha site (Fig. 2.2). Table 2.2 Sex and age of respondents

District Sex Age (years) Total

male female 18-30 31-40 41-50 51 and above

Lalmonirhat 30 10 3 14 12 11 40

Gaibandha 29 11 11 13 8 8 40

Kurigram 26 14 6 13 9 12 40

Total 85 35 20 40 29 31 120

0

5

10

15

20

25

30

35

40

Lalmonirhat Gaibandha Kurigram

Nu

mb

er

Fig 2.2 Education level of respondents

Higher secondary

Secondary

Primary

Illiterate

Household size averaged 5.7 persons in both Gaibandha and Kurigram, but only 4.9 persons in Lalmonirhat. There is a male biased sex ratio on average in the survey households (ranging from 57% male in Gaibandha to 60% male in Kurigram), seasonally migrated household members (mostly men) are included as household members, but the reason for the bias and low number of women is unclear. Overall 55% of sample households have men who migrate seasonally in Lalmonirhat (where most households are protected by embankments), while in Gaibandha and Kurigram (where the households are unprotected and largely live in charlands) just over 80% of sample households have men who migrate seasonally. No women were reported to migrate


seasonally for work from the sample households. Seasonal migration is investigated in detail in Chapter 7. In total 74% of the sample households have labouring as their main source of income, this is consistent across the three districts/study sites. Although most (62% of all sample households) gave agricultural labouring as their main income source, in fact the majority of households do a mixture of agricultural and non-agricultural labouring whenever work is available. Thus 58% of households reported non-agricultural labour as their second occupation and 24% gave agricultural labour as their second occupation. No more than 9% of the surveyed households reported farming (including own land and sharecropped land) as a significant income source. Only 15 of the 120 household surveyed own small areas of cultivable land - mostly in the high risk char areas, and 30 households share crop or lease-in land, again mostly (73%) in the char areas. Investigating the role of community and other local organisations in helping households cope with floods and erosion was one of the aims of the study. Although the study areas were selected for being within the areas covered by local community based organisations (CBOs), very few of the sample households are members of these CBOs, with none in the Gaibandha site. However, over half of the Lalmonirhat sample households and over 10% of households in the other areas access support such as loans from and saving with NGOs whose groups they belong to (Table 2.3). Table 2.3 Membership of NGO groups and Community Based Organisations

District CBO member NGO group member

No hh % No hh %

Lalmonirhat 6 15.0 22 55.0

Gaibandha 0 0.0 4 10.0

Kurigram 4 10.0 6 15.0

Total 10 8.3 32 26.7


3 LALMONIRHAT (MOSTLY PROTECTED MAINLAND) 3.1 Context Three locations in different Upazilas of Lalmonirhat District, all along the Teesta River, were purposively selected. The locations are shown in Fig. 3.1 but embankments (not mapped) have been built along both banks of the Teesta River. In each there is an active CBO known as a federation - a local development association - with a fixed territory, formed under previous development project initiatives. FHRC has facilitated the involvement of these federations in networking for natural resources management in the past. They are cooperatives which pool resources, and can be termed interest associations which admit members from poor strata of the local society. Each Federation operates at union level which is the lowest tier of the local government of Bangladesh. The federations follow a two tier structure – a set of small neighbourhood groups at the lower level and federating these a central committee with an office. The three CBOs are summarised in the following paragraphs. Satata Rajapur Federation operates in Rajapur (or Rajpur) Union of Lalmonirhat Sadar Upazila and covers 22 villages, the Teesta River flows through the middle of the union. In 2016 among the villages here, 60% of Rajapur village and 50% of Thakurdari village were reported to have been eroded, 30% of a further seven villages has been eroded, also 75% of two villages are located in chars within the active channel of the Teesta. In general 70% of remaining land is considered high and 30% low. Progoti Tushvander Federation covers Tushvander (or Tushbhandar) Union in Kaliganj Upazila, on the north side of the Teesta River with much of the union between the river and embankments. Of the nine villages within this area (Tushvander, Sundrahobi, Talikbarinagar, Kanchanshar, Dakshin Ghanasham, Uttar Ghanasham), two were eroded (Kashiram-1 and Bairati-3) a few years back but in 2016 it was reported that recently land here accreted and has been reoccupied, while only Talikbarinagar is completely protected by embankments. Overall about two thirds of land in this area is reported to be high and one third low. Onnesha Velabari Federation covers Velabari (or Bhelabari) Union in Aditmari Upazila, between the Teesta and Dharla Rivers. The 10 villages here (Falimari-1 (part char-adjacent), Falimari-2, Velabari (part char-adjacent), Shalmara (part char-adjacent), Mahishtali (part char-adjacent), Talukdulali, Paglarchara, Pashchim Velabari, Purba Velabari, and Mohammadpur) are on average 80% high land and 20% low. Focus group discussions were held in all of these three areas, and household interviews were conducted in Satata Rajapur Federation - interviewing respondents living largely behind the embankments, but with a minority living in the char areas of that union. Livelihoods differ between the three areas: Satata Rajapur is mainly inhabited by farmers with their own land, with some working as labourers or traders in the town, in Progoti Tushvander only 20% are farmers, but 30% are fishers and half of households mainly depend on work outside the area, also in Onnesha Velabari half the households depend substantially on seasonal migration, and the others mainly work locally, with 15% fishing. In Satata Rajapur during the last 20 years about 5% of households reportedly moved 2 to 5 times due to erosion, but never moved more than 2 miles. Some participants mentioned moving 8-10 times since 1990 but again each time within about 2 miles, and some people moved 2 or even 3 times within the last year (2016).


Fig. 3.1 Lalmonirhat District showing three unions covered by the study


Floods are an annual event here, and 2015 was worse among recent years, particularly in Onnesha Velabari (Table 3.1). However, erosion was reported to be more severe in 2016 and 2015 compared with previous years. Also in Tushvander and Velabari areas morphological changes are reported to be more rapid now than in the past - with chars accreting after a shorter interval in areas that eroded. Table 3.1 Past hazard experience in last 10 years - Lalmonirhat Satata Rajapur

Federation Progoti Tushvander Federation

Onnesha Velabari Federation

Floods Every year but was high in 2015 and in 2011.

Causes misery every year but little more in 2004, 2008, 2011 and 2014.

Flood was more severe in 2015 than in 2016 - water level was 1 ft higher in homesteads and in agricultural field where it stayed for more than a month in 2015. Floods affected the area every year but it was little higher in 2011, 2012, 2013 and in 2014 than other years (except 2015).

Erosion and accretion

Erosion every year but severe in 2015 and 2016

Erosion every year but severe in 2004, 2008, 2015 and 2016. Char formation is faster now than before. People do not have to wait for 30-40 years to get their land back.

In 2016 Ratnai River current was stronger than in 10 years back, erosion and accretion are frequent. It takes 2-3 years to accrete land now in comparison to 30-40 years before.

Drought 2011 and 2013 2003, 2014 and 2015 2014, 2015 and in 2016. Other hazards

Excessive rainfall: 2011, 2016 Now-a-days other natural calamities decreased.

Excessive rainfall: 2001, 2005, 2006 and 2014 Other natural calamities decreased but thunderstorm and fog increased now.

Heavy rain and strong wind in 2016. Thunder storm killed several people and livestock in this area in 2015. Fog in 2012 and in 2015 caused severe damage to the winter crops.

3.2 2016 Floods There were four flood peaks reported in this area in 2016, but the September one was not notable in Satata, the main floods were two in July-August (Table 3.2). Crops and housing were damaged, and in Onnesha some people had to move home several times, dismantling their shacks and building again. Substantial parts of two areas were flooded in 2016, and this was accompanied by erosion particularly in part of Progoti Tushvander. Those households that were flooded deeper in their houses mostly moved for about 15 days during the flood peak- in some cases to India which is nearby and had flood free areas, others to embankments and land protected by embankments (Table 3.2). In two of the areas during the 2016 flood people temporarily moved to higher ground, schools, and mainly the nearby embankments. Respondents said that these places became overcrowded, and they suffered from lack of clean drinking water and poor sanitation, which together caused water borne diseases. There was a lack of space for cooking, and there was no outside assistance with cooked food, and there was even a shortage of space for sleeping. Women suffered from lack of security and privacy. Table 3.2 Summary of 2016 flood-erosion event - Lalmonirhat Satata Rajapur Federation Progoti Tushvander

Federation Onnesha Velabari Federation

Number of households within the site

6000 9000 5256

Land flooded in 2016 60% (10 villages and chars) 40% 25% Households flooded 1200 400 100


Satata Rajapur Federation Progoti Tushvander Federation

Onnesha Velabari Federation

Flood pattern 2016 flood hit 3 times (2 times in July and once in August) and stayed for 5 days. Water level on the agricultural land was about 2.5ft and on the homestead water level was about 1 ft. In 2015 flood water stayed on the land for about two weeks. On the charland the water level was about 6 ft and on the mainland it was 4ft..

2016 flood hit the area 4 times, 2 times in July-August and stayed for 15 days, once in August and once in September and water stayed for 2 weeks. The one in August was worse than the others. Homesteads were under 4 to 4.5ft high water. Erosion was more widespread this year than in last 10 years.

2016 flood came 4 times, 2 big ones in July-August which stayed for 10 days each time and 2 small ones in August-September which stayed for 7 days.

Duration of standing flood water

5-7 days depending on land level

10 to 15 days depending on land level

7-15 days depending on land level

Depth of water Homestead: 1 ft Agric land: 2-2.5ft

Homestead: 4-4.5 ft Agric land: -6-7ft

Homestead: 1 ft Agric land: 5ft

No of households evacuated for flood

460 400 0

Households eroded 800 200 None - homesteads are far from the river bank.

Land eroded 20%. Villages in chars suffered erosion more than mainland. Homesteads eroded more than the agricultural land this year.

100% land of one big village. Part of two newly formed chars eroded.

20%. Usually 5% erodes every year

Movement in flood-erosion event

10 households moved to mainland. 50 households moved to Paglarhat village. About 300 households moved to India. Some moved to Rajarhat and Rajibpur upazila and others moved temporarily to adjacent areas. Poor try to stick to the same area.

200 homesteads eroded: 100 households moved temporarily to embankment and 100 households moved to the protected area. Also all 400 households flooded in their homes moved temporarily within the area, to the embankment, higher ground and shelters, school ground.

People managed to stay at home and did not move

Duration stayed outside the homestead in 2016 during flood

15 days 15 days na

3.3 2016 Flood and Erosion Impacts on Households Among the sample households 12 reported losing some land averaging about 0.52 acres (0.21 ha) to erosion but no other erosion losses, and valued this land loss at over Tk 100,000 per affected household (Table 3.3). However, 70% of surveyed households experienced damage to their houses in these floods, although none were destroyed. Those with land suffered some crop damages, and a minority reported livestock losses. Probably most significantly, all but one of the surveyed households reported on average loosing ten days of paid work due to flood disruption (having to move home in some cases, and being at home to take care of the household in response to floods). Over half of households reported that some household member experienced ill-health during the floods although the average costs of treatment were modest. Overall flood losses in damage and lost income were reported as equivalent to about US$168.


Table 3.3 Summary of 2016 flood and erosion losses to surveyed households in Lalmonirhat Type of loss Impact Flood Erosion No hh mean No hh mean Housing/ homestead loss

Area (dec) 209 total 5.2 mean

0 total 0 mean

No house destroyed 0 0 No house damaged 28 0

Crop/land loss Area (dec) per affected household

22 54 12 52

Livestock Cattle/buffalo no 5 1.4 0 0 Goat no 9 1.6 1 3 Poultry no 16 11.4 0 0

Paid work/ Employment

Days lost/affected 39 10.3 0 0

Health No persons ill 22 1.2 0 0 Days affected 22 6.0 0 0 Costs of treatment (Tk)

22 566 0 0

Total value of loss /amount needed to replace (Tk)

40 13,273 12 117,750

Means are for those households affected by a loss type

3.4 Role of Embankments The embankments along the Teesta River breached in two out of the three areas covered by the study in this district in 2016, but breaches were quickly filled in one site (Table 3.4). Embankments in all three sites provide shelter for the people affected by flood and erosion. Even in the location with several breaches the embankment and groynes are viewed favourably because people say they reduced erosion from what it could have been. However, as is discussed further in Chapter 7, relatively few households in the sample survey study area moved to embankments as temporary shelter during the floods. Table 3.4 2016 Role of embankments along Teesta River in Lalmonirhat Satata Rajapur Federation Progoti Tushvander


Status of embankment

Embankment was breached in 3 to 4 places, despite 3 groynes here, due to strong current.

Eight km embankment was not breached and people living inside the embankment were safe.

The embankment was built in 1989. Several breaches as they river current was very strong. But government and people repaired it.

Role of embankment

Embankment, especially groynes, saved the villages from massive erosion this year. However, a char is forming and the residents hope to have less erosion next year. Embankment is used as shelter during flood and erosion and also used as road for communication. Respondents said it was “inhuman” to live on the embankment as it was so crowded with char and flooded mainland people. Embankment is also used as road during dry season. There is a railway track which is also providing protection.

Embankment saved people and provided shelter to the flood and erosion victims. Due to embankment sand deposition on the agricultural land inside the protected area was avoided. However, water logging from torrential rain impacted people inside the protected area. Erosion victims build houses and live on the embankment.

Embankment works as shelter for the flood and erosion victims in past years.


4 GAIBANDHA (MAINLY RIVERSIDE AND CHARS) 4.1 Context Three locations were investigated, all in Gaibandha district, these locations are in three different Upazilas, covering a combined total of 13 villages inhabited by about 4,300 households and covering just under 4,000 ha. The main sources of livelihood in these areas are fishing, farming (as wage labourers, as share croppers, and for a few households farming own land) and migration for work, in two of the sites 50% or more of households have members who seasonally move away for work. The locations are shown in Fig. 4.1, which also shows the location of the “Brahmaputra Right Embankment” bordering these unions. In each of these three areas a community based organisation has been active for several years (named in the column headings in Table 4.1) working to improve natural resource management and the lives of inhabitants. Notably the site with the largest area reported the lowest number of households (and has had relatively less erosion in the past) and vice versa the smallest of the three areas is by far the most densely populated due to past erosion which has resulted in most people living on public land. These CBOs and their locations comprise: Konai Brahmaputra Community Based Fisheries Organisation in Gazaria Union, Fulchari Upazila covering five villages (Khamarpara (char, eroded), Baluchar, Gobindi, Nilkuthi and Katlamari) with about 60% of land considered high and 40% low. This area includes a branch channel of the Jamuna River (where the fisher members of the community organization fish), parts of island and attached chars, and an area of mainland mostly outside of the embankment. Asher Alo Unnayan Sangstha (1) in Kamarjani Union of Gaibandha Sadar Upazila covering four villages (Karaibari (char) and Goghat 1,2,3 (all facing erosion), with about 30% of land considered to be high and 70% low. Most of the area now comprises of chars and Jamuna River, along with a smaller area of mainland on the riverside of the embankment. Asher Alo Unnayan Sangstha (2) - Vati Bochagari Unnayan Samity in Kapasia Union of Sundarganj Upazila. The union is mostly a series of char islands just downstream of the confluence of the Teesta River with the Jamuna, along with small areas of attached chars outside of the embankment. The community organisation covers four villages (Ujan Bhurail (char), Vati Bhurail (char) Rajarchar (char) and Vati Bochagari). These villages are heavily eroded - out of their original areas 60% of Ujan Bhurail village eroded, 17% of Vati Bhurail village eroded, 20% of Rajarchar village eroded, and all of Vati Bochagari has disappeared. Focus group discussions were held in all of these three areas, and out of them household interviews were conducted in the area covered by Asher Alo Unnayan Sangstha (2) - Vati Bochagari Unnayan Samity in Kapasia Union, the sampled households here either live on lands between the embankment and main river channels (areas that can be termed attached chars) or on the embankment itself having been affected by erosion. These three sites are frequently affected by river floods and by erosion (Table 4.1). Although this varies between years and between locations, the pattern was reported to be more or less the same in Konai Brahmaputra and in Asher Alo 1, while Asher Alo 2 is affected by floods and erosion to some degree each year. The most hazard prone of the three (Asher Alo 2) interestingly reported that morphological changes have been more rapid in the last


decade than in the past - with stronger currents, erosion, and reappearance of land due to accretion.

Fig. 4.1 Study areas in Gaibanda District


Table 4.1 Past hazard experience in last 10 years - Gaibandha Konai Brahmaputra

Community Based Fisheries Organisation

Asher Alo Unnayan Sangstha (1)

Asher Alo Unnayan Sangstha (2): Vati Bochagari section

Flooding Every year but a bit more in 2004, 2005, 2008, 2011 and 2014.

Causes misery every year but a bit more in 2004, 2005,2008, 2011 and 2014

Most years



Erosion every year but severe in 2015 and 2016 Char formation is faster now than before. People do not have to wait for 30-40 years to get their land back.

River current is stronger than 10 years back, river bed silted, erosion and accretion are frequent. It takes 2-3 years to accrete land now in comparison to 30-40 years before.

Drought 2011 and 2013 2011 and 2013 2014 Storms Excessive rainfall: 2011, 2016

Now a days natural calamities decreased.

Excessive rainfall: 2011, 2016 Other natural calamities decreased but thunderstorm increased now.

Heavy rain and strong wind in 2016 and 2013, Thunder storm killed 16 people in this area in 2015-16.

4.2 2016 Floods There were multiple flood peaks in 2016. Although the three sites are not so far apart they reported slightly different patterns, all with the main floods in July-August but in total 3-5 flood peaks, including two in September (Table 4.2). Extensive areas of these three sites were flooded in 2016, and at least half of all households had up to 1.8 m of flood water within their houses for 10-20 days, with their homesteads surrounded by water of around 2-3 m depth (water depths inside houses vary and depend also on house plinth levels). Well over half of households temporarily evacuated during this flood season, and many lost some land to erosion. In Asher Alo 2 all of the flooded households moved temporarily either to shelter on the embankment or at locations protected by the embankment. In Asher Alo 1 erosion was most severe as 400 households lost their homestead land. Most affected people moved temporarily to embankments and higher public buildings for 20-30 days. Table 4.2 Summary of 2016 flood-erosion event - Gaibandha Site/CBO Konai Brahmaputra Asher Alo Unnayan

Sangstha (1) Asher Alo Unnayan Sangstha (2): Vati Bochagari section

Number of households within the site

500 800 3000

Flood pattern Biggest since 1988 (1998 flood did not stay for so long). 2016 flood caused more damage than 1988 flood. 5 flood peaks: early June for 5-6 days. July for 15 days August for 1 month - this was the devastating one as it causes damage to standing crop. September for 18-19 days end September for 5-6 days. Farmers lost crops that had been planted for 2nd time in this last flood.

One foot lower than flood level in 1988. But this year erosion and food hit at the same time making life difficult as they could not return home within a month. Three flood peaks: July/August for 20 days - homesteads were under 5 to 6 feet of water. Even the highlands went under 1-2 feet water. Two peaks in September which stayed for 10-12 days. Erosion was more this year than in last 10 years.

Four flood peaks: 2 big ones in July-August which stayed for 20 days each time 2 smaller ones in August/September which stayed for 10 days. Erosion was greater this year than other years.

Impact of flood Damaged houses, infrastructure, crops (Jute and Paddy), seedbeds were under water

Damaged houses, infrastructure, crops (Jute and Paddy), seedbeds were under water. Lots of sand deposition

Some people had to move several times, dismantling their shacks and building again. They lost all their


Site/CBO Konai Brahmaputra Asher Alo Unnayan Sangstha (1)

Asher Alo Unnayan Sangstha (2): Vati Bochagari section

on the agricultural land. Farmers are afraid that may be in next two years they will not be able to harvest good crop

agricultural land.

Land flooded in 2016 50% 100% 100% Households flooded 250 400 2400 Duration of standing flood water



20 days

Depth of water Homestead: 3-5 ft Agric land: 6-10ft

Homestead: 5-6 ft Agric land: 7-10ft


% of flooded households evacuated due to flood

50 75 80

Households erosion affected (agricultural land or homestead)

230 600 (400 homesteads affected)

750

% area reported eroded 35 40% 52% Movement in flood-erosion event

90 households moved to nearby areas 100 households moved to embankment and flood shelters. Others moved temporarily to adjacent areas.

250 households moved within CBO area to embankment, higher ground and shelters, school ground. 50 households move to adjacent area (will move back when new char appears).

2400 households (all those affected) moved to nearby villages, within protected area, on the embankment and on higher ground. Rest stayed back in their original places, living on macha (raised platform) or raised plinth.

Duration stayed outside the homestead in 2016 flood

1 month 20 days 1 month

4.3 Losses from 2016 Floods and Erosion Among the sample households 18 were affected by erosion, with 14 reporting complete loss of their homesteads (building and homestead land) and four more partial damage to homes from erosion, 15 also lost agricultural land averaging 0.82 acres (0.33 ha) (Table 4.3). A few of the erosion affected households reported loss of livestock in the erosion events and aftermath, but surprisingly very few reported loss of work to erosion. This may be because they lost work during the floods and the time they temporarily evacuated, and did not experience additional employment losses when they found their land had been eroded. Overall the 18 erosion affected households reported associated losses of just under Tk 1 million each - averaging the equivalent of about US$11,600 per household - a considerable loss indicating that within the sample relatively better off households were affected by erosion. A further 20 households reported that their houses were damaged in the floods, but very few owned agricultural land, and so there was little opportunity for crops to be affected (most had already lost any land that they had in past years to erosion). Livestock losses here were significant, with all of the households with flooded homesteads reporting loss of poultry (about eight birds per household), and several reporting loss of larger livestock. A majority (80%) of all surveyed households reported loss of work due to floods (this includes those affected by erosion and as noted above floods and erosion are best considered a combined hazard event when affected people describe impacts on their employment and work). Affected people averaged 22 days of lost work per household due to floods, displacement/evacuation, and time spent coping. Moreover 82% of the households reported


ill-health of at least one person during the floods and spent the equivalent of about US$ 24 per household on treatment. However, this was only a small part of total average flood related losses, considering all 40 surveyed households the average reported value of losses was equivalent to about US$228 per household. Table 4.3 Summary of 2016 flood and erosion losses to surveyed households in Gaibandha Type of loss Impact Flood Erosion

No hh mean No hh mean Housing/ homestead loss


488 total 12.2 mean


Crop/land loss

Area (dec) per affected hh

2 20 15 82

Livestock Cattle/buffalo no 7 1.4 4 2.3 Goat no 11 2.9 3 2 Poultry no 20 7.9 4 10


Days lost/affected 32 22.2 3 40


33 1880 2 9000


40 18,035 18 915,894


4.4 Role of Embankments In addition to damage to crops and housing, in Ashar Alo 1 high sand deposition in floods may affect cropping in the next two years in the lands outside the embankment. In this area in 2016 the Brahmaputra Right Embankment held and protected lands and people living behind it. In the two other areas the embankment breached, adversely affecting the people living behind the embankment. But in all three of these sites many people living normally on the riverside of the embankments moved to take shelter on the embankments when they were affected by flood and erosion (Table 4.4). Table 4.4 2016 flood and role of embankments - Gaibandha Konai Brahmaputra Asher Alo Unnayan Sangstha

(1) Asher Alo Unnayan Sangstha (2): Vati Bochagari section

Role of embankment behind the villages

About 0.5 km of the embankment breached. This big breach in embankment flooded people living inside the embankment. The people living outside the embankment took shelter on the embankment. But respondents said it was “inhuman” to live there as it was so crowded with flooded mainland people as well as char people. Embankment is also used as a road during dry season.

Embankment was not breached and people inside the embankment were safe. Embankment saved people who sheltered on it. Erosion victims built houses and live on the embankment. Due to embankment sand deposition on the agricultural land inside the protected area was avoided. However, flood from torrential rain also impacted people’s life within the protected area.

Embankment breached during the flood in several places. Heavy rain water logging inside, and pressure from outside current caused the breaches. Local people used sand bags and put earth in the breached areas to protect the embankment. Embankment provided shelter for the flood and erosion victims.


5 KURIGRAM (MAINLY CHARS SEPARATED FROM MAINLAND) 5.1 Context Three locations in two Upazilas of Kurigram District were purposively selected. In each there is an active CBO locally termed a federation or samity (society), these are local development associations covering multiple villages within a fixed territory. They were formed under previous development project initiatives, and FHRC has helped these CBOs network with other CBOs to improve their natural resources management in the past. Two of the areas are predominantly in the island chars of the active Brahmaputra River. Goruhara Gram Unayan Samity in Panchgachia Union in Kurigram Sadar Upazila includes nine villages (Kadamtala, Sitaijhar, Balamajhir char, Dakshin Goruhara, Kamarchar, Goldarchar, Dharalar char, Mazhipara, Dhakshin Nowabash) bordered by the Dharla River to the west and Brahmaputra River to the east. About half the land is considered high and half is low. Part of the area is protected by embankments along both the Dharla and Brahmaputra Rivers, and the rest may be considered as attached charland. Jatrapur Federation in Jatrapur Union under Kurigram Sadar Upazila includes 18 villages within the chars of the Brahmaputra River, out of which six villages are entirely in the riverine chars and the other villages are partly in the chars, but also include smaller areas of mainland with a small part of the union located behind embankments. Land is considered to be 65% high and 35% low. Porarchar Gram Unnayan Samity in Begumganj Union of Ulipur Upazila is in the Brahaputra river chars close to the outfall of the Dharla River. It covers seven villages (Mondolpara, Hazipara, Kalimuddinpara, Memberpara, Porarchar, Sluice gate para and Inol Haque Para), and about 60% of land is considered high and the rest low. The area is entirely outside of embankments. Focus group discussions were held in all three of these areas. The FGD in Jatrapur only covered the unprotected char area of the Union. Also the household interviews were conducted in that part of Jatrapur federation in Jatrapur Union which is located within the chars. In Goruhara and Porarchar 50% or more of households were reported to depend on seasonal migration, and the others work as labourers, fish, or farm their own land (about 30% of Goruhara households). In Jatrapur 60% of households are reported to be involved in agriculture and the rest work in towns. All three areas experience flooding every year, but reported it was slightly higher than normal in several recent years, and that 2016 was exceptional (Table 5.1), for example flooding homesteads unlike 1998 or being almost as high as in 1988 (the two largest previous floods affecting the area). Respondents said that flood usually comes here in August-September, but this year it started in July. Erosion is also an annual hazard here, but was worse in 2016. Two sites reported that accretion is more rapid here in recent years. Before it took 15 to 20 years (or even more) for a char to arise and it might be far from the eroded area which influenced violent conflict over access to the new land. Now chars rise up faster, within 5-6 years and near to the eroded char, so it is easier to locate their own land. Respondents said the river is shallower and more braided now.


Fig 5.1 Study areas in Kurigram District


Table 5.1 Past hazard experience in last 10 years - Kurigram Jatrapur Federation Goruhara Gram Unayan

Samity Porarchar Gram Unnayan Samity

Flooding Every year but was not so high. Flood affected in 2010, 2014 and in 2015 due to high rainfall. Flood was 2 ft less than in 1988.

Causes misery every year but more in 2013, 2014, 2015 and 2016. During 1998 flood homesteads were not flooded but continuously there was standing water for 3 months on the agricultural land

Affected every year but it was little higher in 2011, 2012, 2013 and 2014. Flood in 2016 was little less than 1988 flood, but river was violent in 2016. According to the respondents water level was higher in the homesteads in 2016 than in 2015.



Erosion every year but severe in 2013

Erosion and accretion are frequent. In 2016 Dharla river current was stronger than 10 years back.

Drought 2011, 2012, 2013 and 2014 2010, 2011 and 2015 2010 and in 2015/16. Other hazards Excessive rainfall: 2010,

2015 Now a days the period of other natural calamities shifted a month ahead. Untimely rain, drought, fog, shorter winter and less winter rain are evident.

Excessive rainfall: 2014 and 2014. Other natural calamities decreased but thunderstorm and fog increased now..

Fog in high in 2014 caused severe damage to the winter crops.

5.2 2016 Floods All of the unprotected land in the these areas was flooded in 2016 (Table 5.2), and all of the households in two study areas and many of those in the other area were flooded, with water present for 5-15 days in Jatrapur and 15-30 days in the other two areas, and homesteads flooded to a depth of 1-1.5 m. Erosion mainly affected agricultural land, but two complete villages were eroded (close to 400 households) plus 15-20 other households in Goruhara Gram area. These erosion affected households, plus half of flooded households in Jatrapur moved to embankments and higher land for about two weeks, whereas almost all households in Porarchar Gram stayed in their houses living on raise platforms (macha) or boats for two weeks. Table 5.2 Summary of 2016 flood-erosion event - Kurigram Jatrapur Federation Goruhara Gram Unayan


Number of households

5825 2000 960

Area 7000 ha 728 ha 380 ha Flood pattern Flood hit 3 times (once in July, once

in August and once in September). The big one in July stayed for 16 days but in August and September stayed for 5 days.

Flood hit the area 3 times, 2 times in early July to early August and stayed for 20 days, and once in August and water stayed for 2 weeks. The one in early July was worse than the others

3 peaks - one big one in July/August for 15-20 days and two small ones in August-September for 7days. Water stayed for one month on agricultural land.

Land flooded in 2016

66% (12 villages and chars) 100% 100%

Households flooded

<2000 2000 900

Duration of standing flood water

5-16 days 15 to 30 days depending on land level

15-20 days in the homestead and one month on the agricultural land.

Depth of water Homestead: 4-5 ft Agric land: 5-12ft In the unprotected area the water

Homestead: 4-5 ft Agric land: -10-15ft

Homestead: 3-4 ft Agric land: 7-8ft Lowlands: 12ft


Jatrapur Federation Goruhara Gram Unayan Samity

Porarchar Gram Unnayan Samity

level was about 10-12 ft and in the protected area it was 5ft.

No. of households evacuated due to flood

1700 200 14

Impact of flood Damaged houses, infrastructure, crops (Aman, Mash kalai, peanuts), seedbeds were under water. Sand deposition on agricultural land also indicates that within next 2 years they will not be able to harvest better crops.

Damaged houses, infrastructure, crops (Paddy), seedbeds were under water. In exposed areas

Some people had to move several times. They had to dismantle their shacks and rebuild. About 1.5 to 2ft sand has been deposited on all the agricultural land. Some standing crops were buried under sand. They had to sell their livestock.

Land eroded in 2016

25% land eroded. Villages in chars (Garuhara-50%, Parbatipur-25%, Char Jatrapur mouza) suffered more than the villages adjacent to mainland. Erosion happens every year but this year current was high and agricultural land eroded more than homesteads.

100% land of Kadamtala and Dakshin Goruhara village eroded, in other char villages about 15-20 households eroded partly.


Households eroded 27, Within 10km land area in last 10 years only 1.5 km area exists. Now the people living in that area have no land.

In two villages all households eroded and in other chars 15-20 households eroded .

20 (most homesteads are far from the river bank).

Movement in flood-erosion event

1000 households moved to mainland - to embankment, higher ground and adjacent areas. Others made high platform and lived there. Some lived on the roof, some lived on boats

All 400 eroded households moved to the protected area, on the embankment, higher ground and shelters, school ground.

Most people managed to stay at home on raised platform and boats and did not move

Duration stayed outside the homestead in 2016 flood

16 days 20 days 15 days

In all three areas there were the same three flood peaks, with the first in July lasting longest (15-20 days). Overall crops and houses were damaged, with some people in Porarchar having to dismantle and move house; in two areas sand was deposited on land which will affect crops in future years. Erosion resulted in loss of crop land and houses. Those who lost their homesteads (mainly Goruhara Gram) had to pay the costs of moving and stay in open spaces exposed to rain and storms. They had to sell some livestock and became separated from their original community. To cope with erosion and with floods, households borrowed from NGOs, CBO (federation), and moneylenders, moved to cities for work, sold livestock and fixed assets, and caught fish in the river. During the 2016 flood those people who temporarily moved to higher places (schools, embankment, roads) faced overcrowding, shortage of drinking water and latrines, and also places for cooking as no organisation provided cooked food. Lack of safe drinking water commonly caused diseases. Relief was not enough. Women in particular suffered from lack of security and privacy. People in Porarchar preferred to stay on raised platforms in their homestead or on a boat near their homestead due to fear of theft and with the hope that the water would recede quickly.


5.3 Flood and Erosion Losses in 2016 Among the sample households eight were affected by erosion, with seven reporting complete loss of their homesteads (building and homestead land) and one suffering partial damage to its home from erosion, seven also lost agricultural land averaging 1.25 acres (0.5 ha) (Table 5.3). The erosion affected households did not report other types of loss from erosion (although for example they also suffered losses associated with flooding at the same time which are discussed in the next paragraph). Overall the eight erosion affected households reported associated losses of just under Tk 0.5 million each - averaging the equivalent of about US$6,270 per household - relatively low considering the areas of land reported lost due to the low value attached to flood and erosion prone land in the chars. Almost all sample households (35 out of 40) reported that their houses were damaged in the floods. Three houses that were destroyed by erosion suffered flood damage prior to being eroded. Also 42% of households had small areas of agricultural land affected by floods. Livestock losses here were significant, with all of the households with flooded homesteads reporting loss of poultry (about eight birds per household), and several reporting loss of larger livestock particularly goats. All but one of the surveyed households reported loss of work, averaging 23 days of lost work per household due to floods, displacement/evacuation, and time spent coping. Moreover 70% of the households reported ill-health of at least one person during the floods lasting for about 16 days, and spent the equivalent of about US$ 36 per household on treatment. However, this was only a small part of total average flood related losses. Considering all 40 surveyed households, the average reported value of flood losses was equivalent to about US$303 per household. Table 5.3 Summary of 2016 flood and erosion losses to surveyed households in Kurigram Type of loss Impact Flood Erosion No hh mean No hh mean Housing/ homestead loss


184 total 4.6 mean


Crop/land loss Area (dec) per affected hh

17 40.2 7 125

Livestock Cattle/buffalo no 7 1 0 0 Goat no 15 2.3 0 0 Poultry no 35 7.7 1 8


Days lost/affected 39 23 0 0


28 2891 0 0


40 23,949 8 495,225


5.4 Role of Embankments Porarchar has no embankment nearby (which is one reason why people had to stay in their houses on platforms during the 2016 floods), whereas flood and erosion victims moved to the embankments which are not too far away in the other two areas. In Jatrapur federation area normally the embankment is used for shelter during floods and erosion and also it is used as a road. In the adjacent chars are 3 km of embankment of


which 1 km was breached in two places, cutting the road-cum-embankment. As most of the char area is unprotected they could not protect their standing crops (Aman rice, Mashkalai and peanuts). There are also four shelters (higher ground, known as killa) in this area. In Goruhara Gram area the embankment along the Dharla River was built in 1996 and prevented flooding so that people had fish ponds, different crops and a good irrigation system in the mainland there, before they moved to the new chars where they have to move frequently due to erosion. The western part of this embankment was breached in 2010 and the eastern part eroded in 2013; these have not been repaired. The embankment still provides shelter to flood and erosion victims, but it also traps water inside when it rains a lot, sometimes damaging standing crops. Erosion victims build houses to live on the embankment. There is also one shelter in the nearby mainland area.


6 RESPONSES IN 2016 FLOODS 6.1 Household Responses and Coping with Floods and Erosion 6.1.1 Coping actions taken Household heads/representatives interviewed in the sample survey were asked if they had taken up a range of possible actions to cope with the 2016 floods (note that this does not distinguish erosion, and some respondents were affected by both floods and erosion). Evacuation is excluded here as it is investigated in detail in Chapter 7, and none of the households reported selling land to cope. The incidence of different coping actions reported in 2016 was generally similar in the three study areas (Fig 6.1). However, there are some differences in their future expectations of coping actions that they would take (Fig. 6.2).

0

10

20

30

40

50

60

70

80

90

100

% o

f h

ou

seh

old

s

Fig. 6.1 Coping actions in 2016 floods


In all three areas all households changed their eating practices to cope with the floods - reducing the number of meals and amount eaten, and mostly also eating lower quality / less preferred foods as well as limiting how much adults ate. Moreover 70% of households in all three areas borrowed food (in addition to food aid that mostly of them received - see later). In future fewer of the respondents hope/expect to have to do the same if/when they are flooded again, especially borrowing food which carries social stigma and is presumably an indication of a more severe flood/crisis. A minority of households did casual labour for food, or pledged labour for advance payments or sold in advance expected harvests/produce. Compared with the other strategies, migration for work was more widely reported as part of coping strategies - and the seasonal migration for work discussed in Chapter 7 in detail needs to be seen in this context. It is by now a normal component of livelihood strategies, but 30-40% of the surveyed households undertook migration for work specifically to cope


with flooding in 2016 (to recover income and find work when none was available after floods). Migration for work as a coping strategy may have been under-recorded as members of some households may have migrated for work in response to flood impacts but after the survey date, depending on how long men from flood and erosion affected households needed to be with their families. Also in the more flood and erosion prone riverside and char areas of Gaibandha and Kurigram more households expect to migrate for work to cope in future with floods, compared with the Lalmonirhat households who live mostly behind embankments.

0

10

20

30

40

50

60

70

80

90

100

Fig. 6.2. Coping actions households expect to adopt in future


Relatives and friends were by far the most common source of borrowed funds, particularly in Kurigram (70% of households), while NGOs were a source for 40% of households in Lalmonirhat, and 25% of Gaibandha households borrowed from moneylenders. However, livestock sale is one of the notable coping actions as a way to raise money and to avoid deaths and problems facing fodder shortages. In 2016 about 50% of households reported selling small animals to help cope with the floods, ownership of cattle is less common but 20-30% of households sold cattle. In future in all three areas a majority of households expect to sell livestock to cope in future floods, particularly large livestock. This indicates an expected modification of strategy presumably to invest more in cattle when they have funds and to anticipate or respond to floods and sell cattle. As discussed further below this is influenced by availability of grazing in the dry season, opportunities to sell cattle at Eid Ul Azhar, and cattle being a mobile asset. 6.1.2 Self assessments of coping services and strategies The previous section focused on actions taken by households during and immediately after floods, this section reports how respondents rate the value to them of services and features of the socio-hydrological landscape (from social capital to infrastructure) that may or may not have featured in their 2016 coping strategy. The households covered in the sample survey made use of a wide range of services and social connections to help them survive in the


2016 floods. These are summarised in Figs. 6.3-6.5, and the main points of this are highlighted here, followed by more in depth descriptions of coping strategies derived from focus groups. The survey highlighted the importance of local social capital within the village, higher places for shelter, having savings, keeping livestock as a movable adaptable asset, and to some extent the use of mobile phones.

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%T

rustw

ort

hy

rela

tive

s a

nd

neig

hbours

in

vill

age

Tru

stw

ort

hy

pers

on

s in

to

wn

or

city

Em

ban

kme

nt

Pub

lic b

uil

din

gfo

r she

lter

(e.g

.s

chool)

Hig

h land

Kee

pin

gliv

es

toc

k

Savin

gs

Cre

dit

(N

GO

)

Cre

dit

(mo

ney

len

der,

sh

op..)

Mobile

pho

ne

% o

f re

sp

onde

nts

Fig. 6.3 Rating of coping strategies in 2016 flood in Lalmonirhat

Very important

Important

Useful

Little help

No-use

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Tru

stw

ort

hy

rela

tive

s a

nd

neig

hbours

in

vill

age

Tru

stw

ort

hy

pers

ons in tow

n o

rcity

Em

ban

km

en

t

Public

build

ing f

or

shelter

(e.g

.sch

ool)

Hig

h la

nd

Keepin

g liv

esto

ck

Savin

gs

Cre

dit (

NG

O)

Cre

dit

(mon

eyle

nder,

shop

..)

Mobile

phon

e

% o

f h

ou

se

ho

lds

Fig 6.4 Rating of coping strategies in 2016 floods in Gaibandha

Very important

Important

Useful

Little help

No-use

A majority of respondents made use of help from trusted relatives and neighbours in their village and rated this important or useful. Many of the respondents (42%) had no savings to call on during the floods, but for those that had savings these were important in coping. Keeping livestock was important for about two thirds of all households as a way of coping – by selling animals. Livestock coping strategies are discussed further below in Section 6.1.3. Embankments were used/provided benefits that were rated very important by a majority of households in Lalmonirhat and Kurigram – in Kurigram as a safe shelter for char people, and in Lalmonirhat for helping to protect households even if they were eventually flooded in 2016. However, embankments were reported not to have been used or to have been of no use by most households in Gaibandha, this is for a particular reason. The opinion of people in


Gaibandha (and not in the other two study areas) is that as they have to move long term and/or short term to the embankment but get no help there, and because they have to make their own home and get no other services on the embankment they said that it is not useful (even though they actually shelter there). Relatively few households used public buildings for shelter, but those that did (including 17 households in Gaibandha) said this was important or very important to them, and similarly more households in Gaibandha used high land for shelter and considered this important.

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Tru

stw

ort

hy

rela

tives

and

neig

hbo

urs

invill

age

Tru

stw

ort

hy

pers

ons in t

ow

n o

rcity

Em

bankm

ent

Pub

lic b

uild

ing f

or

sh

elte

r (e

.g.

sch

ool)

Hig

h la

nd

Kee

pin

g liv

esto

ck

Savin

gs

Cre

dit (

NG

O)

Cre

dit

(mon

eyle

nd

er,

shop

..)

Mobile

phon

e

% o

f h

ou

se

ho

lds

Fig. 6.5 Rating of coping strategies in 2016 floods in Kurigram

Very important

Important

Useful

Little help

No-use

Only 25% of all households surveyed reported using credit from NGOs to cope with 2016 floods but this was an important component of coping for those that did borrow. Surprisingly only 20% of respondents reported using credit from moneylenders and shops, and only 5% reported receiving help from CBOs when rating its usefulness – but this is inconsistent with the incidence of help reported elsewhere in the same interviews and shown below in Section 6.2. Despite high levels of illiteracy and poverty among the respondents 85% made use of mobile phones during the 2016 floods. Although this was not a very important component of coping, most rated this useful. Mobile phone use did not appear to be part of accessing help from cities and towns, as only 20% of respondents said help from trustworthy people in town was part of their coping. Although ill-health was widely reported in the 2016 floods, this was a greater problem in the unprotected areas particularly the chars. For example, 70% of households lost on average 16 working days per household due to illness in the floods and in addition to that loss spent almost Tk 3,000 per household on treatment (Table 6.1). Table 6.1 Reported health effects of 2016 floods

Lalmonirhat Gaibandha Kurigram No households reporting illness 22 33 28 % households affected 55 82.5 70 No persons 27 45 34 Days income loss from illness per affected household 5.95 11.82 16.50 Cost of treatment (Tk per person) 461 1,379 2,381 Cost of treatment (Tk per household) 565 1,880 2,891

Only 2 households in Kurigram reported health effects of erosion


6.1.3 Livestock In all three areas cattle are widely kept and form an important movable asset. Ownership of cattle and associated losses and challenges in moving and maintaining cattle may have been greater in 2016 than in other years because in 2016 the Muslim festival of Eid-ul-Azha took place in September at the end of the monsoon season, and many households in this region decided to fatten cattle for sale as prices are usually high for the festival and in that year supply was expected to be low as India restricted cattle movement to Bangladesh. Usually in the dry season it is easy to graze and maintain livestock in the extensive grassy areas of the chars making this a valuable enterprise, and people also keep cattle as a safety net. In the monsoon cattle are stall fed, or sold off. In all three areas in the 2016 floods cattle were moved to higher ground such as school grounds, embankments and other higher land and remained there for up to a month, until the homesteads dried up. In Lalmonirhat a few livestock died due to shortage of feed and proper shelter, or disease linked with crowding. Although middlemen tried to pay less knowing their inability to sell cattle by themselves during flood, supply was low compared with demand and respondents were happy with the prices. In Gaibandha in the relatively less flood and erosion prone areas there were greater cattle losses as more households still had cattle expecting to profit from sale in September for Eid ul Azha. About 30% of the livestock died due to poor feed, lack of proper shelter, or disease. In Ashar Alo 1 despite problems providing feed when cattle were moved to higher places, but only about 5% died and most people could sell their livestock for the Muslim festival. Whereas Ashar Alo 2 is more flood and erosion prone and few households keep any livestock in the monsoon (they sold them earlier) as they have to move several times in a year. In Kurigram half of the respondents fatten cattle and sell them during or before the monsoon. Very few livestock died. However, they said they could not get proper prices as middlemen paid less knowing their inability to sell cattle by themselves due to floods and also the lack of shelter. In one of the sites about 30% of households sold cattle to recover after the flood, and because it was difficult to feed and keep cattle healthy during floods. 6.2 Assistance Received 6.2.1 Role of different bodies From the household sample surveys relatively few households received help from the CBOs in their area (consistent with low CBO membership in the sample). This was mainly in the form of warnings of the flood. CBOs were the main source of warnings in Kurigram and Gaibandha (Fig. 6.6). A majority of flood affected households received food aid from the government during the floods. In Lalmonirhat most also were warned by the government about impending floods. Relatively few households received assistance with immediate shelter, although almost 50% in Gaibandha were helped by government agencies with shelter. Subsequently relatives and friends were important sources of help with rebuilding houses.


0

5

10

15

20

25

30

35

Warning Shelter Housing Food Warning Shelter Housing Food Warning Shelter Housing Food

Lalmonirhat Kurigram Gaibandha

Nu

mb

e r

of

sa

mp

le h

ou

seh

old

s

Fig. 6.6 Number of households reporting sources of assistance in 2016

NGO

GO

CBO

Relatives/friends

6.2.2 Value of different types of support In the household survey respondents were asked to rate the effectiveness of different types of assistance if they had received them in helping to cope with floods and erosion. Figs 6.7 to 6.9 summarise the findings. In general early warning of floods was the most valued assistance in all three areas with the vast majority of respondents rating this good or very good. Knowledge and understanding of hazards was also rated as helpful (good or very good) by 70% or more of respondents in Lalmonirhat and Kurigram but under 50% in Gaibandha. Similarly information on relief and work opportunities was considered more useful in Lalmonirhat and Kurigram. Protection of homes and land from floods and erosion was rated highly in all three areas, as were flood free locations for shelter. Somewhat surprisingly a majority of households rated micro-credit as of no use in coping, but opinions were more polarised on this with other households rating this a good form of assistance. The responses may be influenced by a mixture of lack of access by some households to micro-credit, NGO/lender rules that limit use in the context of disasters, and concern among flooded households not to fall into debt. Agricultural support was also considered as of no use by a majority of respondents, but this is because most do not have own land or grow crops themselves. Unlike credit, help received to reconstruct houses after floods and to move home during floods and erosion was highly rated by almost all households in the three areas. Similarly cash help and food aid were rated highly in all three areas, indicating appreciation of practical help from government and other sources during the flood and evacuation period. More than half of respondents rated health care in support of coping good or very good, but many households considered health care of only average value. However, there was very little appreciation of law and order indicating a failure of these authorities to help secure flood affected people’s homes and property either in their origins or their evacuation locations.


0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Ho

useh

old

s

Fig. 6.7 Rating of help to cope with riverine hazards in Lalmonirhat

Very good

Good

Average

Poor

No use

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Ho

us

eh

old

s

Fig. 6.8 Rating of help to cope with riverine hazards in Gaibandha

Very good

Good

Average

Poor

No use


0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

% o

f h

ou

seh

old

s

Fig 6.9 Rating of support to cope with riverine hazards in Kurigram

Very good

Good

Average

Poor

No use

6.2.2 Role of CBOs Despite the modest level of help from CBOs found in the sample survey, in focus group discussions CBOs were found to have played important local roles, although this was limited by their modest resources. In all three locations in Lalmonirhat the CBOs are significantly involved in disaster related activities - providing warnings, helping rescue and move people and their possessions, helping affected people access relief and services from local councils, government and NGOs, and also contributing some materials for the worst affected (Table 6.2). The CBOs form a link with NGOs and other agencies that supported flood warning dissemination through community volunteers in two areas. Because the CBOs here are federations with links to NGOs they have relatively wide coverage and put a priority on disaster risk reduction related activities. In Gaibandha NGOs are active in all three areas, providing general advice, flood warning systems with community volunteers in two areas, and in the most erosion prone area where two villages have eroded seven times NGOs, mainly Practical Action have helped people move house, raise house plinths and establish new homes, and provided cattle (Table 6.3). Schools in addition to embankments are used for shelter, and in all three areas government provided food as relief. CBOs differ in their capacities and services in floods, but even in Konai Brahmaputra where this is an organisation of poor fishers it formed a basis for sharing warnings and for members helping one another to move and rebuild house. In all three locations the CBOs have formed a basis for voluntary assistance in disasters - helping to warn and rescue people, and also contributing some materials for the worst affected.


Table 6.2 Support for coping in 2016 in Lalmonirhat Satata Rajapur Federation Progoti Tushvander Federation Onnesha Velabari

Federation Initiatives to help people

Early warning was given by different NGOs, Government and Union Parisad. However, people saw the river current and realized that there will be disaster. They moved to higher ground in the main stable area.

Early warning by NGO and Government. Volunteers were trained before on preparedness and raised awareness among people. There is a Village Disaster Management Committee. Besides hoisting flag, warnings were spread by loudspeaker (miking) by the volunteers. Embankment and schools were used as shelter. Government provided some relief although people complained it to be not enough.

The CBO members did not get receive much help as there was very little damage. They got help from the federation.

CBO role during flood/ erosion

One of the main activities of the CBO is disaster management. The CBO members received training and formed volunteer group. They provide early warning through own miking system. They help relocate people by using their boats and labour, find space for the people and their livestock to move. They report to RDRS (a regional NGO) for relief and help. They also keep contact with government for post flood/erosion rehabilitation. They also cooperate with Union Parisad during relief distribution.

They provided information on flood and erosion to their members They formed rescue team and helped people to move to safe places with boat and provided some help for a day. They prepared list of flood and erosion victims, appealed to NGOs and government agencies for help for the victims. They also communicate with the Union Parisad. They also help people to rebuild their houses and repair roads. Distributed seeds, tubewell, latrine and medicine among the erosion and flood victims.

The CBO has a Disaster Management Committee. They provided information on flood and erosion to their members. In 2016 they provided early warning and helped vulnerable households to safer places before flood. They also provided information on affected people of the area and distributed relief provided by the government and INGOs.

Table 6.3 Support for coping in 2016 - Gaibandha Konai Brahmaputra Asher Alo Unnayan

Sangstha(1) Asher Alo Unnayan Sangstha (2): Vati Bochagari Unnayan Samity

NGO and government initiatives to help people

Early warning was given by different NGOs, Government Information Services, Red Crescent, Social welfare department and Union Parisad. However, schools were used as shelter and people moved by themselves as everyone was affected and very difficult to help each other. Government and NGOs provided dry food for one day only.

Early warning by NGO and Government was provided. Volunteers were trained before on preparedness and 40 (20 from young group and 20 from community) received training from NGO. Besides hoisting flag, volunteers gave warning by loudspeaker (miking). Embankment and schools were used as shelter. Government provided relief (rice, dal salt).

In 2016 two villages Vati Bochagari and Rajar char was eroded three times. Since 1988 char started to erode and since then up to 2015 these chars eroded 7 times. People moved to the land area with the help of the NGO. NGO provided information on flood and erosion. Practical Action, an INGO, raised 100 house plinth and renovated house, built latrine and provided tubewell. In 2007 erosion victims who moved to embankment received cattle. Government provided relief (rice, dal salt).

CBO role during flood/ erosion

CBO members are also very poor. Most fish for living. They helped people to move and when needed they helped them to build shelters. Members collect bamboo from whoever has a bamboo grove to build shelter on embankments. They also provided early warnings to their members and their trained volunteers gave training on how to move to a safer place, how to preserve food and essentials, etc.

CBO members helped people to move, provided some help for a day. They provided dry food, formed rescue team and helped people to move to safe places with boat. They also provided information on the flood and erosion to their members. Distributed seeds among the erosion and flood victim.

CBO provided information on the flood and erosion to their members. They also distributed relief provided by the government and INGOs.


In Kurigram in Porarchar area no NGO or government assistance was reported, and despite almost all houses being flooded people did not move away and reported little damage. There their CBO helped warn people and to move some vulnerable households. In the other two areas also CBOs were active helping people to move and rebuild their houses or do repairs, and also forming a link with NGOs and government, including local councils, to access relief for the worst affected by erosion or floods. Also NGOs and government besides relief gave flood warnings in these two areas, although people said they could also tell for themselves that it would be a severe flood (Table 6.4). Table 6.4 Support for coping in 2016 - Kurigram Jatrapur federation Goruhara Gram Unayan


NGO and Government Initiatives to help people

Early warning was given by different NGOs, Government and Union Parisad. Looking at the river current and rainfall upstream, char people realized that there would be a disaster. They moved to higher ground in the stable area.

Early warning was provided by NGO and Government. Embankment and schools were used as shelter. Government provided some relief although people complained it was not enough.

The CBO members did not get much help as there was very little damage.

CBO role during flood/erosion

One of the main activities of the CBO is disaster management. The CBO members received training and formed volunteer group. They provide early warning through own miking system .They relocate people by using their boats and man power, find places for the people and their livestock to move. They report to RDRS (a regional NGO) for relief and help. They also keep contact with government departments for post flood/erosion rehabilitation. They also cooperate Union Parisad during relief distribution.

CBO members helped people to move and build new houses or repair damaged ones. But as the members are all poor and have very limited funds they were not able to do much. They prepared lists of flood and erosion victims, and appealed to NGOs and government agencies for help for the victims. They also communicated with the Union Parisad.

The CBO provided information on the flood and erosion to its members. They provided early warning and helped vulnerable households move to safer places by CBO boat before flood. They also provided water purification tablets and tablets for diarrhoea.


Table 7.1 Main advantages and disadvantages reported for home location (% of households, note multiple responses possible so each % is from those who evacuated)

Factor Lalmonirhat Gaibandha Kurigram Total Advantage

Relatives and known people 38.5 46.4 38.5 40.6 Safe location 5.1 28.6 25.6 18.9 Productive land 23.1 7.1 23.1 18.9 Good communications 12.8 21.4 12.8 15.1 Good public services (school or health etc) 12.8 3.6 10.3 9.4 Good environment 7.7 3.6 12.8 8.5 Land to live on 2.6 7.1 0.0 2.8 Plenty of work 7.7 0.0 0.0 2.8

Disadvantage No/poor sanitation 33.3 71.4 17.9 37.7 Poor communications 28.2 7.1 33.3 24.5 Risky location 7.7 21.4 30.8 19.8 Don't get enough work 17.9 28.6 7.7 17.0 Theft/crime/tolls 2.6 14.3 15.4 10.4 Poor/no accommodation 15.4 14.3 2.6 10.4 Poor food 10.3 3.6 2.6 5.7 Poor public services (school, health, etc) 2.6 0.0 10.3 4.7 Poor environment 2.6 0.0 2.6 1.9 People not helpful 0.0 0.0 2.6 0.9 Unproductive land 0.0 0.0 2.6 0.9 No land to live on 0.0 0.0 2.6 0.9 Far from family/worry 0.0 0.0 2.6 0.9 Total households evacuated 39 28 39 106

7 MIGRATION AND MOVEMENTS 7.1 Home Location In all three areas the main reason respondents gave for choosing the places where they regularly lived before the 2016 floods was that they own land there (Fig. 7.1). Access to land (not own land) in that area, and having relatives nearby were the other main factors. In Gaibandha another reason given was being close to their home (original own land considered to be home by people who had lost their homes in the past from erosion). Very few respondents stated that they were living in their birthplace or that that was the reason for their location (the majority of the surveyed households have over the respondents’ lifetimes moved home one or more times). When people who evacuated in the 2016 floods were asked to compare their home location with places that they had relocated to (see Section 7.3), social capital in the form of relatives and other known and trusted people was reported as the main advantage of living in their home locations in all three districts (Table 7.1). After this safety, followed by communications and environment were highlighted in Gaibandha and Kurigram; whereas the other main attraction or advantage of home locations in the Lalmonirhat area is productive land. The

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%


Fig. 7.1 Main reason given for location of home

safe from flood

work available

near home

land available

relatives here

own land here

birthplace


main disadvantage of their home location reported especially in Gaibandha (where more people live on embankments) was poor sanitation, and only in Kurigram for original home was sanitation considered a lesser problem. Poor communications and risks were considered more significant disadvantages in the Kurigram char areas, after these a wide range of disadvantages were reported by small proportions of the households. 7.2 Incidence of Different Types of Movement The majority of surveyed households had to leave their home during the 2016 floods, particularly in Lalmonirhat and Kurigram (Table 7.2). In addition a majority of households have at least one member who moved away from home for part of the year for seasonal work - particularly in Gaibandha and Kurigram where sources of livelihood are limited, of these six households in Gaibandha and two in Kurigram had a second person who migrated seasonally for work. However, only 20% of households moved and did not return to their starting home in 2016 (the year reported is actually November 2015 to October 2016), with fewest permanent relocations in Kurigram and most in Gaibandha (for example households that had shifted due to erosion and were living on the embankment at the time of the survey). Nevertheless over the longer term of a respondent’s lifetime permanent relocation of home is quite common as discussed in Section 8.1. Table 7.2 Numbers of surveyed households involved in different types of movement in 2016

District Evacuation Permanent move Seasonal (for work) no % no % no %

Lalmonirhat 39 97.5 8 20.0 22 55.0

Gaibandha 28 70.0 9 22.5 33 82.5

Kurigram 39 97.5 6 15.0 30 75.0

Total 106 88.3 23 19.1 85 70.8

7.3 Short-term Hazard Response Movement in 2016 (Evacuation) One of the widely documented and investigated responses to floods globally is temporary evacuation when homes are inundated and become uninhabitable. To this can be added the complexity of movements also influenced by erosion - which can mean that what was expected to be a short term evacuation is extended, or that people move back to a location close to but not exactly where they had lived. Almost all of the surveyed households, except for those already living on embankments in Gaibandha, had to leave their homes in the 2016 floods, moving from their respective main homes to take temporary refuge in higher places. In the vast majority of cases evacuation was because their homes were flooded, although nine sample households reported they were eroded, and two moved because they found better places to live (Table 7.3). In Lalmonirhat and Kurigram most moved to the nearest embankments and a few to villages behind the embankments (Fig 7.2) whereas in Gaibandha those evacuating moved to other places either on the riverside of the embankment (presumably higher than their homes) or behind the embankment. Most respondents (94%) said they chose the place they moved to because it was safe from floods, just six households selected places mainly for being close to relatives or because they owned land there.


Table 7.3 Reason for evacuating in 2016 (number of respondents)

District Eroded Flooded Found better place to go Total

Lalmonirhat 0 38 1 39

Gaibandha 6 22 0 28

Kurigram 3 35 1 39

Total 9 95 2 106

Fig. 7.2 Locations of sample households before and during evacuation 2016

Location lived in 2016 before evacuation

0

5

10

15

20

25

30

35

40

45


Nu

mb

er

of

ho

useh

old

s

village area behindembankment

on embankment

riverside (not protected)

island char

Location evacuated to in 2016

0

5

10

15

20

25

30

35

40

45


Nu

mb

er

of

ho

us

eh

old

s other more distant ruralarea

village area behindembankment

on embankment

riverside (not protected)

island char

Note: excludes households that did not evacuate (e.g. those already living on embankments)

Out of 106 surveyed households that evacuated in 2016, 26 (25%) after returning to their homes had to evacuate a second time due to flooding. The first evacuations took place mainly in July in Gaibandha and Kurigram, but in August in Lalmonirhat (Table 7.4) when mainland areas near the Teesta River were flooded. Table 7.4 Month of first evacuating from home in 2016 (% of respondents)

District June July August No households

Lalmonirhat 0.0 0.0 100.0 39 Gaibandha 3.6 89.3 7.1 28 Kurigram 0.0 66.7 33.3 39 Total 0.9 48.1 50.9 106


Not only the timing, but also the durations of evacuation differed between the three study areas: not only was evacuation later in Lalmonirhat but also for a much shorter period of 1-10 days, and only two households out of 40 evacuated twice (Table 7.5). By comparison along the Brahmaputra in Gaibandha and Kurigram the typical duration of evacuation was 22-30 days for the first event and an additional 16-30 days for those households that evacuated twice. The vast majority (92%) of respondents left their evacuation location when water receded and they could go back home, the other 8% (those whose homes were eroded) found better places to move to or were forced to move again by landowners. Table 7.5 Duration of evacuation from main homestead in 2016 (% of households)

District 1-5 days 5-10 days

11-15 days

16-21 days

22-30 days

31-60 days

No. Households

First evacuation Lalmonirhat 61.5 35.9 0.0 0.0 2.6 0.0 39 Gaibandha 0.0 3.6 10.7 17.9 64.3 3.6 28 Kurigram 0.0 2.6 15.4 28.2 48.7 5.1 39 Total 22.6 15.1 8.5 15.1 35.8 2.8 106 Second evacuation Lalmonirhat 100.0 0.0 0.0 0.0 0.0 0.0 2 Gaibandha 0.0 0.0 18.2 45.5 27.3 9.1 11 Kurigram 0.0 0.0 7.7 15.4 76.9 0.0 13 Total 2 3 7 13 1 26

Respondents almost all reported that the places they moved to were safe from flood, and thus the main advantage they reported for their evacuation locations (Table 7.6) differed considerably from their origin (main home), see section 7.1. In all cases the main advantage of evacuation locations was safety from floods. Table 7.6 Main advantage reported for evacuation locations (% of households)

Advantage Lalmonirhat Gaibandha Kurigram Total Relatives and known people 7.7 0.0 10.3 6.6 Safe location 79.5 82.1 87.2 83.0 Productive land 0.0 0.0 0.0 0.0 Good communications 12.8 7.1 0.0 6.6 Good public services (school or health etc) 0.0 0.0 2.6 0.9 Good environment 0.0 3.6 0.0 0.9 Land to live on 0.0 3.6 0.0 0.9 Plenty of work 0.0 3.6 0.0 0.9 Total households 39 28 39 106

Similarly very few of the respondents considered their evacuation locations to be risky whereas this was the third most commonly cited disadvantage about their home (origin) before the floods (Table 7.7 compared with Table 7.1). Notably the main disadvantage reported for the evacuation location was poor or no sanitation - usually considered a concern of women but here emphasised by both men and women respondents. In common in all three areas in addition to poor sanitation in the evacuation locations the other main problem was food - expressed as poor food but relating to both quantity and quality.


Table 7.7 Main disadvantages reported for evacuation locations (% of households, note multiple responses possible so each % is from those who evacuated) Disadvantages Lalmonirhat Gaibandha Kurigram Total No/poor sanitation 46.2 75.0 64.1 60.4 Poor communications 2.6 0.0 2.6 1.9 Risky location 0.0 3.6 0.0 0.9 Don't get enough work 7.7 7.1 10.3 8.5 Theft/crime/tolls 2.6 3.6 15.4 7.5 Poor/no accommodation 10.3 3.6 5.1 6.6 Poor food 30.8 35.7 30.8 32.1 Poor public services (school, health, etc) 5.1 0.0 2.6 2.8 Poor environment 5.1 3.6 2.6 3.8 People not helpful 5.1 0.0 0.0 1.9 Unproductive land 2.6 0.0 5.1 2.8 No land to live on 0.0 0.0 0.0 0.0 Far from family/worry 0.0 0.0 0.0 0.0 Number of households evacuated 39 28 39 106

Consistent with these patterns a majority of respondents rated their home locations moderate to high risk, or even high risk in Gaibandha (Table 7.8), whereas evacuation locations were generally rated low risk, except in Kurigram where more respondents considered they evacuated to high or very high risk locations because the embankment itself is now threatened by erosion. Table 7.8 Respondents' assessments of risk in their origin and evacuation locations (% of households)

District Low Moderate High Very high Total Origin Lalmonirhat 7.7 43.6 48.7 0.0 39 Gaibandha 3.6 14.3 60.7 21.4 28 Kurigram 12.8 41.0 33.3 12.8 39 Total 8.5 34.9 46.2 10.4 106 Evacuation location Lalmonirhat 71.8 25.6 0.0 2.6 39 Gaibandha 71.4 10.7 17.9 0.0 28 Kurigram 38.5 7.7 38.5 15.4 39 Total 59.4 15.1 18.9 6.6 106

7.4 Long-term Hazard Response Movement 7.4.1 Lalmonirhat People who have money and are affected by erosion in this area were reported in the focus groups to have bought land outside the area and left permanently in the past, but the percentage is very small. Some households went to India and never came back. In the focus groups some cases of women who left to work in cities in garment factories were reported and they can only come home during festivals or long holidays as they only get one day off weekly, and also may work overtime to make more money. In the focus group discussions it was reported that very few people migrated permanently as they wanted to stick together with their family and be secure and get help from each other. They said they are afraid of moving from their birth place and they do not have money to buy better land outside the area and they do not know the people elsewhere. Nevertheless eight (20%) of the sample households moved home, even though their homesteads were not eroded or destroyed in 2016. This was because they consider they were at future risk and decided to move a short distance in advance of anticipated erosion.


7.4.2 Gaibandha According to the focus group in Konai Brahmaputra area a few people who have money bought land outside the area and left permanently, but the percentage is very small - still about 30 households were reported to have migrated to Dinajpur District (also in northwest but far from the main rivers) some time in the past and never came back. However, in the other two areas few if any households were reported to have moved far away as they prefer to stay close to their kin and get help from each other as well as from NGOs in the area. They are afraid of moving from their birth place, and they do not have money to buy better land outside the area. In addition 14 of the sample households (35%) had their homes destroyed by erosion and nine of these relocated their households permanently as a result in 2016, while remaining in the same area. However, the pattern in this area is to move a short distance to the embankments or other higher places, in the expectation that they will be able to move back to either the homes they evacuated from (short-term), or (medium-term) to land that emerges in the chars close to where they lived. 7.4.3 Kurigram In Jatrapur area only a few people who have money were reported to have bought land outside the area and left permanently, or went to India and never came back. But in this area, compared with the other two sites, more people were said to have made local movements due to erosion. According to the focus groups, in the last 10 years 5% of households were reported to have moved 2 to 5 times but remained within the adjacent area. Some participants in the focus group said they moved home 10-12 times since 1990 but never moved more than 2 miles. And some people moved 2 to 3 times within the last one year. In Porarchar Gram no households were known to have migrated permanently from this area, but 20% of households have members who moved to cities long-term to work in garment industries (see next section). These women only come home during festivals or long holidays. Goruhara Gram area is significantly different – no households migrated permanently as they said they want to stick together with their family and be secure and get help from each other. They said they are afraid of moving from their birth place and they do not have money to buy better land outside the area and they do not know people in other areas. 7.5 Livelihood Related Movement 7.5.1 Patterns of seasonal migration In the focus groups it was reported that ten years back fewer people were migrating and when they did they went mainly to adjacent towns and areas such as Rangpur and Bogra; but now more people move and they go to more distant places where they can get work. ln all three areas of Lalmonirhat it is common for men to migrate seasonally for work - from at least 50% of households, the timing depends on availability of work - some is seasonal for planting and harvesting crops in other districts, and others are away for much of the year working in factories, or for shorter periods when they have less work on their own land (Table 7.9).


Table 7.9 Seasonal migration for work - Lalmonirhat Satata Rajapur Federation Progoti Tushvander


% of households involved

75% household male members 60% household male members from unprotected area and 40% from the protected area

50% household members from unprotected area move temporarily every year for

Urban work Dhaka and Chittagong (construction work, brick field, pulling rickshaw, professional work like carpentry, masonry)

Dhaka and Chittagong (construction work, brick field, pulling rickshaw,)

15% go to Dhaka and adjacent area to work in the garment industry.

Agricultural work Comilla, Noakhali, Bogra. Mushiganj for planting and harvesting potato.

Comilla, Tangail and Bogra.

Bogra and Sherpur area to plant and harvest Aman paddy.

Duration of temporary/ seasonal migration each time

Depends on types of work. They usually come home every 3 to 4 months but usually stay out of their home for 6 months. Those who have agricultural land come home frequency to help family. Those who go to work as agricultural labourers go on a contract and in a group. Usually they go each year to the same area they are familiar with.

On average 4 to 6 months a year. Urban labourers go for 30 to 60 days, come home and again go. Those doing agricultural work go twice, once for sowing/ transplanting and once for harvesting. Those who work in garment or other factories go for longer. They usually come for holidays.

Mostly 4-6 months a year for those who go for agricultural work and wage labour (construction, brick field etc.) Those who pull rickshaws work for 15-21 days at a time and come home for a week or two and then they again go to cities.

Other factors For coping they also sold land, took loan from NGO and from Federation. They also work overtime to earn more money.

20% of the households catch fish in monsoon season and only 10% area involved in risky agriculture.

A high proportion of households in the Gaibandha study areas are involved in seasonal migration for work (see Table 7.10). Although there are some differences between the three sites, the general pattern is that men are away for much of 6-8 months during the dry season taking any work they can get for survival. Some of this work is on trips of 1-2 months at a time, other men are away for the full period. Especially in the most flood-erosion prone site, but also in the others, men come back to their original home/family during the monsoon mainly so that they are there to help their family move or cope with floods and erosion.


Table 7.10 Seasonal migration for work - Gaibandha Site/CBO Konai Brahmaputra Asher Alo Unnayan Sangstha

(1) Asher Alo Unnayan Sangstha (2): Vati Bochagari section


25-30% 70% 75%

Timing October-June, typically 6 months

6-8 months in dry season, work trips are 30-60 days at a time in most cases, except for fishers who can be away up to 6 months. Some stay away year-round except for holidays

Dry season, at least 6 months, up to 8 months

Urban work Dhaka, Chittagong, work as rickshaw/van puller, earth cutting labour, construction labour, stone carrier. Some people work in factories and live in Gazipur, Joydebpur, Kalagachi and Mirpur.

Dinajpur, Thakurgaon, Rangpur, Dhaka, Munshiganj and work as wage labour in brick field, Garments, drive rickshaw/van/autorickshaw

Most of the people migrate to Dhaka and Chittagong to work as labour in brick field, pull rickshaw and work as construction labour

Agricultural work In Mushiganj area they work specially for planting and harvesting potato. In Sylhet area they go for rice harvesting.

Many men and some women go to Mushiganj for planting and harvesting potato. Others go to Dinajpur, Bogra and Rangpur for agricultural work twice a year, once for sowing/ transplanting and once for harvesting.

They go to Manikganj and Bogra to harvest rice. Migration to Mushiganj is for planting and harvesting potato. 10% of people go to adjacent areas (Sreepur, Chadipur) for harvesting Jute and maize.

Other work The fishers (about 40% of households) fish in groups in distant parts of the river and stay away for 6 months a year (they do some fishing from home here in river in monsoon)

Men from between 60% and 75% of all households in the three areas of Kurigram studied move seasonally away for work (Table 7.11). The patterns are very similar in the three areas, they spend 6-9 months a year away, but come back home every 2 months or less (usually under a month away at a time for rickshaw pullers). Most people go to large cities (Dhaka and Chittagong) for construction work, to work in brickfields or pulling rickshaws. A small number have moved away longer term to work in factories, mainly garments. Those who work away as agricultural labourers go on a contract and in a group to districts southeast of Dhaka to work on planting and harvesting. Usually they go to the same area every year as they are familiar with the land owners and the area.


Table 7.11 Seasonal migration for work - Kurigram Jatrapur federation Goruhara Gram Unayan



60-70% 75% 75%

Urban work Dhaka and Chittagong (garments, construction work, brick field, pulling rickshaw, professional work like carpentry, masonry)

Dhaka(construction work, brick field, pulling rickshaw).

10% go to Dhaka and adjacent area to work in the garment industry. Most go seasonally to Dhaka, Chittagong (construction work, brick field, pulling rickshaw).

Agricultural work Comilla and Noakhali, (agricultural labour).

Comilla, Feni , Munshiganj and Narayanganj Districts

Comilla, Feni , Munshiganj districts to plant and harvest Aman paddy and Potato

Duration of temporary migration

About 6 months. Depends on types of work. They usually come home every 2 to 3 months. Those who have agricultural land t come home frequently to help..

On average 6 to 8 months. Labourers work for 30 to 60 days, come home and again go. Those who go for agricultural work go twice, once for sowing/transplanting and once for harvesting. Garment and other industry workers go for longer, but they usually come home for holidays.

6-9 months. Those who go for agricultural work and wage labour (construction labour, brick field etc.) usually work for 4-6 months a year. Those who pull rickshaws work for 15-21 days and come home for a week or two and then they again go to cities.

Other coping actions

For coping they also sold land, livestock, took loan from NGO and from Federation. They also work overtime to earn more money.

20% of the households catch fish in monsoon season and but 75% are involved in risky agriculture as sand is deposited on farm land every year making land less fertile.

7.5.2 Details of migration for work from sample survey In the households covered by the sample survey all migrants were men. While there could be some stigma and tendency to not report women who moved away for work, we believe this is not hidden in the households surveyed as from discussions women rarely move away for short term work and none of the households surveyed reported any migrants with jobs such as in garment industries. A majority of those migrating temporarily/seasonally for work are aged 16-30 years (Table 7.12), and most are either illiterate or have only primary education (Table 7.13) - limiting their work opportunities. Table 7.12 Ages in years of all temporary/seasonal migrants in sample survey (numbers)

District Less than 16 16-30 31-40 41-50

More than 50 Total

Lalmonirhat 1 9 7 4 1 22

Gaibandha 1 14 9 5 4 33

Kurigram 0 15 7 6 2 30

Total 2 38 23 15 7 85

Table 7.13 Education level of seasonal migrants (numbers)

District Illiterate Primary Secondary Higher secondary Total

Lalmonirhat 8 12 2 0 22

Gaibandha 16 11 6 0 33

Kurigram 16 8 5 1 30

Total 37 27 12 1 77


Most of the individuals reported in the household survey moved to towns and cities for work. Although in the focus groups the same set of destinations came up, the household survey revealed major differences in the proportions of destinations by origin district/location (Table 7.14). From Lalmonirhat people scattered for work but with a third going short distances to nearby districts in the northwest for agricultural work and a similar proportion going to other districts for this type of work (Table 7.15). From Gaibandha 61% moved to Dhaka and its neighbouring industrial areas for urban work, and of those that reported their type of work in Dhaka area most are rickshaw pullers or labourers. While from Kurigram the majority (70%) went to other distant areas - almost all for agricultural labouring work (Table 7.15). Reasons for these differences include the more vulnerable locations (living on embankments) and lack of natural resource based livelihood opportunities among the Gaibandha households as well as shorter distances from that district to Dhaka and its surrounding industrial areas. There are also differences between the three areas in contacts that have built up – the Kurigram respondents have known contacts for agricultural labour in their destination districts. By comparison, agricultural opportunities are better at home and risks are less for the Lalmonirhat households, so those that move go shorter distances and can more easily come home for local work. Table 7.14 Destinations of seasonal migrants (number of persons)

Destination Lalmonirhat Gaibandha Kurigram Total

Northwest region

Bogra 6 2 0 8

Rangpur 1 0 0 1

Dinajpur 0 1 0 1

Sub-total 7 3 0 10

Dhaka area

Dhaka 6 12 5 23

Gazipur 1 4 2 7

Narayanganj 1 2 1 4

Keraniganj 0 2 1 3

Sub-total 8 20 9 37

Other areas

Munshiganj 2 1 13 16

Sylhet 0 1 3 4

Chittagonj 2 4 0 6

Tangail 1 0 2 3

Noakhali 2 0 0 2

Bikrampur 0 0 2 2

Manikganj 0 2 0 2

Narshindi 0 1 0 1

Comilla 0 1 0 1

Mymensingh 0 0 1 1

Sub-total 7 10 21 38

Total 22 33 30 85


Table 7.15 Types of work undertaken by seasonal migrants by origin and destination (% of those reporting)

Location Type of work Lalmonirhat Gaibandha Kurigram Total

NW

Agricultural labour 27.3 11.1 0.0 11.7

Fish trade 4.5 0.0 0.0 1.3

Sub-total 31.8 11.1 0.0 13.0

Dhaka area

Rickshaw pulling 18.2 25.9 7.4 16.9

Non-agricultural labour 4.5 18.5 7.4 11.7


Fish trade 4.5 0.0 3.7 2.6

Fishing 4.5 0.0 0.0 1.3

Other employee 0.0 3.7 0.0 1.3

House servant 0.0 0.0 3.7 1.3

Carpenter 4.5 0.0 0.0 1.3

Petty trade 0.0 0.0 3.7 1.3

Sub-total 36.4 55.6 29.6 41.6

Other


Non-agricultural labour 0.0 3.7 3.7 2.6

Rickshaw pulling 4.5 7.4 0.0 3.9

Sub-total 31.8 33.3 70.4 45.5

Total Number of persons 22 27 27 77

As described in the focus groups, there is some marked seasonality in the timing of migration for work - few people from the surveyed households are away in July and August, particularly from Kurigram considering that migration is common there (Fig. 7.3). With limited work in the study area in those months in 2016, this was so that men could be present/ returned to their homes to help cope with floods and erosion, particularly in the chars. There was also a peak in migration for work in September-October 2016 as part of flood recovery, while peaks in the January to April period align with dry season agricultural work - sowing and harvests. Those that do go away for work tend to spend on average 20-25 days a month working and this did not vary much between months during 2016 (Fig. 7.4).

Fig. 7.3 Seasonality of migration for work

0

5

10

15

20

25

30

Nov Dec Jan Feb Mar Apr May June July Aug Sept Oct

Nu

mb

er

of

mig

ran

ts f

rom

sam

ple

ho

us

eh

old

s Lalmonirhat

Gaibandha

Kurigram


Fig. 7.4 Days worked per individual involved in seasonal migration

0

5

10

15

20

25

30

Nov Dec Jan Feb Mar Apr May June July Aug Sept Oct

Me

an

day

s p

er

mo

nth

Lalmonirhat

Gaibandha

Kurigram

Most of the seasonal migrants made repeated trips for work - mainly to the same places, or in a few cases to different locations in the year, and typically made 3-5 return trips, with some making 1-2 trips (Table 7.16). Table 7.16 Number of return trips made home by seasonal migrants in a year (% of persons reporting)

District None Once Twice 3 to 5

times 6 to 8 times

8 to 10 times

Total no. persons

Lalmonirhat 0.0 36.4 18.2 36.4 4.5 4.5 22 Gaibandha 21.7 26.1 30.4 52.2 13.0 0.0 23 Kurigram 6.7 13.3 16.7 43.3 20.0 0.0 30 Total 7.8 22.1 16.9 40.3 11.7 1.3 77

7.5.3 Issues and hazards faced by migrants In the household survey the respondents reported a wide range of disadvantages and problems faced with seasonal migration for work, although these are weighed against the benefit of earning some income. These were broadly similar for the three origin areas and for the different destinations and to some extent types of work. The five main disadvantages that are common are: poor sanitation in the places they live when away from home, lack of or poor accommodation in those locations, the worry of being far from family, poor diet when they migrate, and high travel costs to get from home to places of work (Fig. 7.5). Broken down by origin, over half of migrants considered that their situation when they go away for work is moderate to very high risk, particularly for people from Kurigram (Fig. 7.6)


Fig. 7.5 Main disadvantage of seasonal migration destination

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%


others

take loan on high interest

high travel cost

poor food

poor/no accommodation

no/poor sanitation

theft/crime/tolls

don't get enough work

don't get paid properly

far from family/worry

no land to live on

poor communications

people not helpful

poor public services (school or healthetcrisky location

little work

Fig. 7.6 Perceived risk in seasonal migration locations

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%


% o

f m

igra

nts

very high

high

moderate

low

none

The issues and challenges faced by seasonal migrants were reported to be similar in focus groups from all three districts and have been consolidated here. Urban migration As there are very few opportunities for work in their home areas, many men have to go seasonally to the cities and even to other places to earn money to maintain their families. All of the migrants complained about the quality of accommodation and food they eat, lack of clean drinking water, and lack of toilets. Most men who move to urban areas for work tend to stay in unhygienic insecure places. Men working in cities usually sleep in crowded or public places, mosques, park benches, footpaths, railway station, bus station, under the shade of


shops in local markets, etc. Those who are construction workers often sleep in the site. Those who come with family often end up sleeping in roadside shacks which they have to dismantle during the day time. Those who earn a better wage (e.g. from Konai Brahmaputra area) can afford to rent space in a room shared by 4-5 people. Getting work every day is uncertain for wage labourers. Construction labourers complained about not getting their wages on time and that they never get the full payment. Rickshaw pullers said they have to do hard work and they cannot continuously work for many days. Often they fall sick and weak. But at least they can earn for the family, if they stay in the village they cannot earn anything locally. Those who pull rickshaw/van sleep on the roadside in shacks made of plastic. Other challenges facing migrants in cities are theft and corruption. For example, rickshaw pullers have to make payments to the police, and street vendors have to pay local mastans (musclemen) and police. Migrants also mentioned the problem of high and increasing travel costs to and from their urban working places. Also sometimes people fall in a trap and get involved in drugs and/or crime. Agricultural migration Agricultural labourers move as a group each year to areas where they previously got work, and stay there until the seasonal work ends. Although getting regular work can be uncertain, there is some security once a group gets a work contract. The agricultural labourers have to find places to sleep near their contract sites. If the contract for agricultural labour includes food they have the advantage of getting food, but mostly for other work they have to cook themselves, the same as for non-agricultural work. In the dry season this work is available but in the monsoon this work is not available and they have to come back home to help their families. Garment workers In the case of garment workers, the main problems are job security and personal security. These people get less salary than the others. They face problems to rent houses which usually cost more than they can afford. So they are forced to stay in low cost houses and in crowded places where they are not connected to water supply and gas. Impacts on women Seasonal migration for work almost always involves men, leaving women living at home to take care of the children and the family, livestock, etc. In these times women have to face different problems (health, food, etc.) all by themselves. Shortage of money is common when their male partners do not send money on time. Some families send their school going children for work to cope with the situation. In the most erosion affected of the sites surveyed (in Gaibandha) no NGO gives loans as they are considered "floating people" and have no permanent address or property. Here the women and children tend livestock in the dry season that are mostly sold before the monsoon as they do not have permanent places to live and lack security for livestock in the monsoon. Debt All households where men migrate have to borrow money in advance to give to the family before they leave. Some households fell in the trap of moneylenders (mahajan) who give loans on high interest rate and then take all the money they earn. The CBOs (e.g. federations) active in the study areas have their own credit system which helps members


who can borrow from the CBO. Also NGOs provide short term loans. But many migrants have to take advances for their work and then must fulfil their job at any cost. The absent men are always worried about the family they left at home, although some men who migrate alone get married again and never come back to the village. 7.6 Role of Communication in Seasonal Movement Although some households did not have anyone moving temporarily in the last year, all of the surveyed households consider this livelihood strategy and so all were asked about their communication methods used to get information on potential work and to communicate with family members while away or for the household to communicate with their absent member. Only two methods were reported - face to face contact and use of mobile phone, others such as letters and text messages were not used. For sources of work, mobile phones were slightly more important than face-to-face contact for all three areas (Table 7.17) - highlighting how important this technology has become even for poorer households such as those covered in this study. Communication with family was mainly face-to-face although mobile phones were also used by almost all of the households, and as a main communication method by many. Communication with others was less regular and both face-to-face and mobile phones were used sometimes. Table 7.17 How households communicate with distant workplaces (% of all households)

Purpose Use

Face-to-face Mobile phone

Lalmonirhat Gaibandha Kurigram Total Lalmonirhat Gaibandha Kurigram Total

Source of work

Not used 7.5 0 0 2.5 7.5 0 0 2.5 Used sometimes 32.5 25 27.5 28.3 20 17.5 15 17.5

Mainly used 60 52.5 57.5 56.7 62.5 62.5 70 65.0

With Family

Not used 0 0 0 0.0 10 0 0 3.3 Used sometimes 7.5 20 7.5 11.7 32.5 17.5 22.5 24.2

Mainly used 92.5 77.5 92.5 87.5 42.5 60 62.5 55.0

With Others

Not used 0 0 0 0.0 5 5 2.5 4.2 Used sometimes 75 52.5 62.5 63.3 52.5 45 57.5 51.7

Mainly used 25 40 37.5 34.2 32.5 30 37.5 33.3


8 TRENDS AND CHANGES OVER TIME 8.1 Lifetime Movements Although few of the surveyed households moved home permanently in 2016, several did lose their homesteads. However, despite being a high flood year 2016 was only one year and local erosion is not necessarily correlated with exceptional floods. Considering the respondents' lifetimes (on average 40 years old, range 20s to 60s), none of them had stayed in the same homestead location since birth (Fig. 8.1). Somewhat surprisingly, since their current homes are in the most stable/least erosion prone locations, the respondents in Lalmonirhat have moved most in their lifetimes - 85% have moved five or more times. This is because most of these respondents came originally from further upstream along the Teesta where they suffered repeated erosion and frequent moves, until they moved to the area that was covered in the survey which is less erosion prone, so most of their shifts in home took place many years earlier. The pattern in the other two areas is more expected - in Gaibandha most households (50%) have moved home 2-4 times, whereas in Kurigram 75% of households have moved 2-8 times, which is consistent with the Kurigram site being further from the mainland in the active chars where erosion and accretion of chars is a regular phenomenon.

Fig. 8.1 Number of times moved home in respondent's lifetime

02468

101214161820


Nu

mb

er

of

ho

us

eh

old

s

1

2-4

5-8

9-20

8.2 Hazards in Past Five Years The vast majority of surveyed households in all three study areas reported experiencing losses due to floods in the past five years, with most experiencing damage to housing and crops in two out of five years, loss of work in three of the years, and ill-health in 1-2 years in five (Table 8.1). Table 8.1 Reported incidence of flood losses in past five years

Type of loss

% households experienced Mean number of years in 5 years

Lalmonirhat Gaibandha Kurigram Lalmonirhat Gaibandha Kurigram

Land/crop 85.0 62.5 80.0 1.97 2.08 2.09

Livestock 80.0 92.5 82.5 1.47 1.78 1.85

Housing 92.5 97.5 97.5 1.92 2.00 1.90

Work 100.0 97.5 95.0 3.03 3.31 2.66

Serious Illness 87.5 92.5 85.0 1.11 1.86 1.32


By comparison fewer households experienced erosion losses in the past five years, although the majority still suffered loss of land, and in Gaibandha 82% of households lost homes to erosion in the past five years (Table 8.2), losses of other types are less common with erosion as in general households can shift movable assets when the river bank is close to their home. Most affected households experienced erosion losses once, or in some cases twice in the past five years. Table 8.2 Reported incidence of erosion losses in last five years

Type of loss



Land/crop 65.0 87.5 72.5 1.50 1.26 1.28

Livestock 30.0 57.5 20.0 0.08 1.26 1.13

Housing 50.0 82.5 42.5 0.80 1.48 1.18

Work 50.0 62.5 27.5 1.05 1.88 1.36

Serious Illness 32.5 40.0 10.0 0.31 1.38 1.00

In the focus groups sand deposition was mentioned as a problem for agriculture in some locations and years, and close to 30% of the surveyed households in Lalmonirhat and Kurigram reported this affected their crop land, typically in two of the last five years (Table 8.3). This affected few households in Gaibandha (as most lost their land to erosion). A small number of respondents also said sand deposition affected livestock rearing because it covered grazing areas. Table 8.3 Reported incidence of losses to sand deposition in last five years

Type of loss



Land/crop 32.5 15.0 27.5 2.23 2.67 1.64

Livestock 2.5 2.5 0.0 2.00 1.00 .

Housing 0.0 0.0 0.0 . . .

Work 0.0 0.0 2.5 . . 1.00

Serious Illness 0.0 0.0 0.0 . . .

Waterlogging (rainwater or floodwater being trapped for longer in an area) was only reported to be significant by about a quarter of Kurigram households. They experienced this on average once in the last five years presumably associated with prolonged floods (Table 8.4). Table 8.4 Reported incidence of waterlogging losses in last five years

Type of loss



Land/crop 5.0 7.5 27.5 3.50 1.00 1.27

Livestock 0.0 0.0 12.5 . . 1.00

Housing 0.0 0.0 27.5 . . 1.45

Work 0.0 0.0 17.5 . . 1.29

Serious Illness 0.0 0.0 10.0 . . 1.25

The common opinion or perception of surveyed households in all three areas is that the frequencies of the two main hazards, floods and erosion, are both increasing (Fig. 8.2). This helps explain expectations of continued or increasing migration for work. Erosion in particular was reported to be increasing in Gaibandha. Sand deposition and waterlogging were reported to be increasing by over 25% of Kurigram households but are largely unknown to households in the other two sites.


Fig. 8.2 Reported trend in hazards in past five years

0%

20%

40%

60%

80%

100%

Flo

ods

Ero

sio

n

Sa

nd

de

po

sitio

n

Wa

terlo

gg

ing

Flo

ods

Ero

sio

n

Sa

nd

de

po

sitio

n

Wa

terlo

gg

ing

Flo

ods

Ero

sio

n

Sa

nd

de

po

sitio

n

Wa

terlo

gg

ing


Ho

use

ho

lds

No answer

Decreasing

About same

Increasing

8.3 Changes in Incomes In the household survey respondents were also asked about the trends in the past five years in their income and expenditure from different possible sources. Very few are involved in some activities: out of 120 households only four in fishing, and four in trade or business. About 70% are or were involved in farming in the past five years, with most in Lalmonirhat where they reported a largely positive trend (Fig. 8.3). However, in the other two areas a smaller proportion of those farming reported increasing income and about 20% of households reported declining income from this source, but at the same time spending on cultivation was reported to have increased or stayed the same so profits would appear to have fallen on average.

Fig 8.3 Trend in agriculture income and expenditure

0%10%20%30%40%50%60%70%80%90%

100%

Inco

me

Exp

en

ditu

re

Inco

me

Exp

en

ditu

re

Inco

me

Exp

en

ditu

re

Inco

me

Exp

en

ditu

re

Lalmonirhat Gaibandha Kurigram All

% o

f h

ou

seh

old

s

Increased

Same

Decreased

Not applicable

Most surveyed households keep some livestock. As has been shown in Section 6.1.3, this is an important component of hazard coping, although they also suffer losses in floods. The majority of households reported increasing incomes from livestock (although less common in Kurigram) with costs either staying constant or also increasing (Fig. 8.4), and from the significance of this component of livelihoods overall returns are presumed to be increasing.


Fig. 8.4 Trend in livestock income and expenditure

0%10%20%30%40%50%60%70%80%90%

100%

Incom

e

Expe

nd

itu

re

Incom

e

Expe

nd

itu

re

Incom

e

Expe

nd

itu

re

Incom

e

Expe

nd

itu

re


% o

f h

ou

se

ho

lds

Increased

Same

Decreased

Not applicable

Most (80-90%) of surveyed households are involved in wage labour in their area, and generally reported that incomes have been increasing while any associated costs have stayed the same (Lalmonirhat) or also increased (Fig. 8.5).

Fig. 8.5 Trend in wage labour income and expenditure

0%10%20%30%40%50%60%70%80%90%

100%

Incom

e

Exp

enditure

Incom

e

Exp

enditure

Incom

e

Exp

enditure

Incom

e

Exp

enditure


% o

f h

ou

seh

old

s

Increased

Same

Decreased

Not applicable

Lastly 55-70% of households said they are or have been involved in seasonal migration for work in the past five years. Almost all reported that their income from this sources has increased (except for a minority of households in Gaibandha), but costs for most households have also increased (Fig. 8.6).


Fig 8.6 Trend in migration income and expenditure

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Inco

me

Exp

en

diture

Inco

me

Exp

en

diture

Inco

me

Exp

en

diture

Inco

me

Exp

en

diture


% o

f h

ou

seh

old

s

Increased

Same

Decreased

Not applicable

8.4 Livelihood Changes Respondents in the household survey were asked to categorize changes in indicators of their livelihood status comparing the present with five years ago. Figs 8.7 to 8.13 show the results reported as number of households, out of all 40 sampled in each of the three locations. The vast majority of respondents in all three areas consider the location they live in now to be worse than five years earlier, which reflects hazards experienced and in some cases moves made (Fig. 8.7). Yet over 50% in Gabandha and Kurigram have improved their house quality in the same period (Fig. 8.8). Almost all reported a worsening landholding situation in the past five years - through erosion and pressure of poverty, or other factors affecting land productivity (sand deposition, flooding) (Fig. 8.9).

Fig. 8.7 Changes in location living in last 5 years

0

5

10

15

20

25

30

35

40


Better

Same

Worse

Much Worse

Not applicable


Fig. 8.8 Changes in house structure in last 5 years

0

5

10

15

20

25

30

35

40


Much worse

Worse

Same

Better

Much better

Fig. 8.9 Changes in landholding in last 5 years

0

5

10

15

20

25

30

35

40


Same

Worse

Much Worse

Not applicable

Most households have maintained or increased the number of days that they have work in over the past five years, including through seasonal migration for work (Fig. 8.10), and 50% or more also reported increasing wages in this period (although inflation means that changes in real incomes may not be large) (Fig 8.11). These opinions are consistent with the overall trend in income from wage labouring reported earlier.

Fig. 8.10. Changes in number of days worked in last 5 years

0

5

10

15

20

25

30

35

40


Much Better

Better

Same

Worse

Much Worse

Not applicable


Fig. 8.11 Changes in wages/income in last 5 years

0

5

10

15

20

25

30

35

40


Much Better

Better

Same

Worse

Much Worse

Despite increases in days worked and wages, 80% or more of households in all three areas reported lower savings than five years earlier (Fig. 8.12). Their savings were run down recovering from the 2016 floods as well as investment in housing and in obtaining work.

Fig. 8.12 Changes in savings in last 5 years

0

5

10

15

20

25

30

35

40


Better

Same

Worse

Much Worse

Not applicable

The end result of these different changes is a mixed opinion from the respondents on their ability to cope with risks and hazards (Fig. 8.13). In Gaibandha, where more households now live on the embankment and savings have fallen, the majority of households consider their coping ability has fallen. In the other two sites less than half of households think their coping ability is worse now. In Lalmonirhat, where most of the respondents live in mainland areas behind embankments, despite 2016 flooding 40% of respondents think their coping ability has improved.


Fig. 8.13 Changes in ability to cope with hazards in last 5 years

0

5

10

15

20

25

30

35

40


Better

Same

Worse

Much Worse

Not applicable

Consolidating these reported trends by study location highlights the respondents' perceived decline in their living standards on most indicators in all three areas (red shades in Figs. 8.14 to 8.16). In Lalminorhat their perceived ability to cope with hazards is reported to have improved for more households (40%) and is in line with improvements in wage income despite worsening savings and landholdings.

Fig. 8.14 Changes in living condition in last 5 years in Lalmonirhat

0%

20%

40%

60%

80%

100%

Hou

se stru

ctur

e

Day

s of

wor

k a

year

Wag

e/inco

me

leve

l

Savin

gsLa

nd

Loca

tion

living

in

Ability to

cop

e with

risk

s

Ho

us

eh

old

s

Much Better

Better

Same

Worse

Much Worse

By comparison less than 10% of Gaibandha households think their ability to cope with risks and hazards has improved even though they reported at least as positive trends in wage incomes and also housing (Fig 8.15). This is consistent with Fig. 8.13, and the perceived decline in coping ability is believed to result from the precarious condition many of the respondents here find themselves in, living on or near the embankment.


Fig. 8.15 Changes in living condition in last 5 years in Gaibandha

0%

20%

40%

60%

80%

100%

Hou

se stru

ctur

e

Day

s of

wor

k a

year

Wag

e/inco

me

leve

l

Savings

Land

Loca

tion

living

in

Ability to

cop

e with

risk

s

Ho

use

ho

lds

Much Better

Better

Same

Worse

Much Worse

Lastly, in Kurigram over half of respondents consider their coping ability is the same or improved in the last five years (Fig. 8.16), but this is despite a very similar pattern of other changes in assets and incomes to that reported in Gaibandha. This may be because of the continued scope for these households to move back to their charland homes after floods.

Fig. 8.16 Changes in living condition in last 5 years in Kurigram

0%

20%

40%

60%

80%

100%

Hou

se stru

ctur

e

Day

s of

wor

k a

year

Wag

e/inco

me

leve

l

Savings

Land

Loca

tion

living

in

Ability to

cop

e with

risk

s

Ho

us

eh

old

s

Much Better

Better

Same

Worse

Much Worse


9 OVERALL FINDINGS AND CONCLUSIONS 9.1 Significance and Impact of 2016 Floods The study focused on representative communities living in the active floodplain of major rivers in three districts of northwest Bangladesh. In all of the three areas investigated the 2016 floods were among the worst experienced there and came in several flood peaks. The majority of surveyed households were affected in some way by flooding in 2016, in addition a substantial minority suffered erosion of their homes. Most households lost income as there was less work available locally and they were busy coping with floods, and many houses (up to 50%) were damaged in the floods with widespread loss of poultry and some other livestock. Average losses per household were about US$168 in Lalmonirhat, US$228 in Gaibandha and US$ 303 in Kurigram. A majority of households (88%) temporarily evacuated their homes during the floods. 9.2 Role of Embankments Many of the communities studied live in char areas or on the river side of existing embankments and as such are not protected by embankments. Yet these households do have embankments relatively close to them, only one of the nine study areas had no nearby embankments. In six out of eight locations with embankments there were breaches in that locality in 2016, raising questions about their intended effectiveness. Nevertheless embankments were widely appreciated as places of refuge, and adverse impacts were not reported. Many households that lost their homes to erosion or had flooding inside their homes moved temporarily to shelter on embankments. Embankments were the main destination of the surveyed households that evacuated in Lalmonirhat and Kurigram. The situation in Gaibandha is more complex, many of the surveyed households already lived on the embankment having lost their homes in the past to erosion, but the embankment is itself threatened by erosion and those households that were flooded in 2016 evacuated to other high land and public buildings. Living conditions in places of shelter (embankments, schools and other high places) were reported to be very difficult due to overcrowding, poor water and sanitation, and difficulty cooking. People from one char location preferred not to evacuate and stayed to protect their homes on boats or raised platforms. Nevertheless, this highlights a different perspective from some highly flood and erosion prone people regarding hydro-social landscapes. Water resource engineering has conceived of structural flood mitigation as enabling risk reduced or even hazard free agriculture and livelihoods. In reality embankments are an important structural resource for coping by inhabitants of the charlands, and serve this purpose even when they are not continuous. At the same time, for inhabitants of the protected area the embankments provide the same coping facility when they breach in high floods but do not ensure a risk-free environment. This raises a question whether government structural responses to flood and erosion hazards in the riverine landscape should focus as they have on expensive rehabilitation and enlargement of long embankments along the main river, or on high shelter points above high return period (exceptional) floods connected by moderately raised but flood resilient embankment-roads that could protect crops in “normal” and low return period floods. Costs saved by this strategy could be used to improve services for urban migrants.


9.3 Role of Local Institutions Warning systems appear to have been quite effective in 2016 with NGOs and government providing flood warnings in the study sites. A finding not noted in past studies is that in the study areas, which were selected for the presence of CBOs, these CBOs played an important role. Particularly in some of the most vulnerable areas, such as chars, the CBOs helped to disseminate warnings, helped the worst affected people move their possessions and salvageable house materials, and helped find safe places for them. They also assisted affected people to access relief and services from NGOs and government. However, as found in past studies, the main sources of immediate help were government, neighbours and relatives, and NGOs. 9.4 Coping Strategies The study found many of the strategies and responses of flood and erosion affected households to cope with these hazards have not changed over time. A mixture of self-help and resilience borne of necessity combined with dependence on government and NGO relief during and immediately after the floods was a common pattern. But in addition to the role of CBOs as intermediaries improving access of their affected members to relief services, the networks and linkages of char and floodplain people with more distant urban and rural areas has altered coping. Not only are households living in the unprotected chars linked with the hydro-social landscape of flood protected areas through finding shelter on embankments and other protected public spaces, they are also more closely linked through employment than in the past with areas where hazard exposure is different, including urban areas. Households in these areas may on average have more options for helping themselves to cope, individually or in groups, than in the past; but there are strong constraints on this resilience. Migration has costs as well as benefits (see below), and households may at best manage to recover their lives after floods and erosion since the vast majority reported that savings had declined in the past five years. Moreover the future prospects for these households are uncertain as flood hazards are perceived to be increasing in frequency. In particular the cumulative impacts of successive unusual floods on coping and livelihoods have not been investigated. Many of the households surveyed reported suffering from floods in 2014 or 15 as well as in 2016, and already 2017 appears to have been a more extreme flood at least for much of the study area, so the cumulative impacts and responses deserve study. 9.5 Role of Migration Three types of movement were recorded. Temporary evacuation was widely adopted by flooded households to move to safer locations - for up to 10 days along the Teesta River in Lalmonirhat, and for 15-30 days along the Brahmaputra-Jamuna River in the other two areas. Some households had to evacuate twice. Permanent movement in the 2016 floods was a response to erosion of homes and was short distance usually to embankments or other nearby places in order to stay close to relatives and be on hand for the future chance of land re-emerging from the river. Hence most households keep a physical and social base in the riverine landscape (on embankments, behind them, or in the chars and unprotected areas according to opportunities and social links). Seasonal migration for work is an important part of livelihood strategies for about 70% of households surveyed, and was also important for many as a way of earning money to cope after the floods. Migration for work, mostly by men, is very common in all of the study sites, particularly in the more hazardous areas. This is either to cities and their edges for work in


construction, labouring, pulling rickshaws, and rarely in factories; or to distant districts as teams/groups to work planting and harvesting crops during peaks in demand for agricultural labour. Especially from the more vulnerable locations, this is seasonal migration for six or more months of the year, usually in several trips, and these men avoid migrating for work in the flood season to be ready to cope and safeguard their homes. Conditions faced by temporary/ seasonal migrants in the places they go for work are difficult - they have no proper shelter, face poor sanitation and drinking water, and are vulnerable to exploitation in the cities. Moreover, they have to borrow to provide initial funds for their wives and families that remain behind, to cover travel costs, and to cover initial living costs when they move; all of which reduce the actual benefit these households receive. 9.6 Implications of Findings In the past many of the people in the study areas could not find work when opportunities in their locality were reduced, for example by floods. Improved communications including the Jamuna Bridge connecting this northwest region with the rest of the country and mobile phones, as well as contacts developed over time through seasonal migration for work, have changed their coping opportunities. Often temporary evacuation during floods as well as seasonal and coping migration for work involves groups of households and individuals. While this has developed spontaneously, CBOs have shown that they respond to hazards and are helping local people organise to improve their coping. There is scope for CBOs, assisted by NGOs and government, to further help vulnerable households and their members to work together in enhancing this strategy. Possibilities include:

accessing services and resources including public lands; strengthening skills to obtain better paid work, in enhancing living conditions for

migrants; technology now offers the scope to support groups and CBOs that are long range

and dispersed seasonally, to improve information sources and contact even flows of funds between households and their members that remain part of riverine and char communities but are scattered whether by erosion or for work;

providing group loans to enable households to fund seasonal migration (including the living costs of families left in the riverine areas) that involve less exploitative terms; and

helping women left behind (when men migrate) organise in parallel groups to men in order to develop income sources and strengthen cooperation to protect their homes and families.

Lastly, since embankments will continue to be used as places of shelter by people affected by floods and erosion, parts of them that are relatively accessible from the chars and to mainland lines of communication should be designed for use as shelter areas from the outset.

Hydro-Social Deltas Migration and hazards 2016 R-1

REFERENCES Ali, M.Y. (1997). Fish, Water and People. (Dhaka: University Press Ltd.). Baqee, A. (1998). Peopling in the land of Allah jaane. Power, peopling and environment: the case of char-lands of Bangladesh. The University Press Ltd, Dhaka. Benson, C. and Clay, E.J. (2002). Bangladesh: Disasters and Public Finance. (Washington, DC: World Bank. Disaster Risk Management Working Paper Series 5). Boyce, J.K. (1990). “Birth of a Megaproject: Political Economy of Flood Control in Bangladesh.” Environmental Management 14(4): 419-428. Brammer, H. (1999). Agricultural Disaster Management in Bangladesh. Dhaka: University Press Ltd. Brammer, H. (2000). “Controversies surrounding the Bangladesh Flood Action Plan.” In: Parker, D.J. (ed.) Floods (vol. 1). (London: Routledge), pp. 302-315. Brammer, H. (2004). Can Bangladesh be Protected from Floods? Dhaka: University Press Ltd. EGIS (2000). Riverine Chars in Bangladesh Environmental Dynamics and Management Issues. Dhaka: Environment and GIS Support Project for Water Sector Planning (EGIS) and The University Press Limited. Elahi, K., Ahmed, K.S. and Mafizuddin, M. (1991). Riverbank erosion, flood and population displacement in Bangladesh. Jahangirnagar University Riverbank Erosion Impact Study, Dhaka. Elliott, J. R. and J. Pais (2006). Race, class, and Hurricane Katrina: Social differences in human responses to disaster. Social Science Research 35(2): 295-321. Falk, W., Hunt, M.O. and Hunt. L.L. (2006). Hurricane Katrina and New Orleanians' Sense of Place - Return and Reconstitution or "Gone with the Wind"?" Du Bois Review 3(1): 115-28. FAP 3.1. (1992). Jamalpur Priority Project Study Char Study Report. Flood Plan Coordination Organisation, Ministry of Irrigation, Water Development and Flood Control, Dhaka (prepared by Sogreah/Halcrow/Lahmeyer and associates). FAP 14 (1992). Flood Response Study Draft Final Report and Planning Guidelines. Flood Action Plan component 14. Dhaka: Flood Plan Coordination Organisation, Ministry of Irrigation Water Development and Flood Control (prepared by Irrigation Support Project for Asia and the Near East). FAP 16 (1995). Meghna Confluence Charland Socio-economic RRA. Flood Action Plan component 16. Dhaka: Flood Plan Coordination Organisation, Ministry of Irrigation Water Development and Flood Control (prepared by Irrigation Support Project for Asia and the Near East).

Hydro-Social Deltas Migration and hazards 2016 R-2

FAP 16/19. (1993). Charland Study Overview: Summary Report, Flood Plan Coordination Organisation, Ministry of Irrigation Water Development and Flood Control, Dhaka (prepared by Irrigation Support Project for Asia and the Near East). de Graaf, G., Born, B., Uddin, A.M.U. and Marttin, F. (2001). Floods, Fish and Fishermen. (Dhaka: University Press Ltd.). Haque, C. E. and Zaman, M.Q. (1989). Coping with Riverbank Erosion Hazard and Displacement in Bangladesh: Survival Strategies and Adjustments. Disasters 13(4): 300-313. Hossain, A. N. H. A. (2006). “The impact of floods on Bangladesh and options for mitigation.” In Siddiqui, K.U. and Hossain, A.N.H.A. (eds.) Options for Flood Risk and Damage Reduction in Bangladesh. (Dhaka: University Press Ltd.) pp 55-70. Hughes, R., Adnan, S. and Dalal-Clayton, B. (1994). Floodplains or Flood Plans? (London: International Institute for Environment and Development and Research Advisory Services). Miah, M.M. (1988). Flood in Bangladesh. (Dhaka: Academic Publishers). Paul, B.K., Rahman, M.K. and Rashit, B.C. (2010). Post-cyclone Sidr illness patterns in coastal Bangladesh: an empirical study. Natural Hazards 56: 841–852. Penning-Rowsell, E., Sultana, P. and Thompson, P. (2013). The ‘last resort’? Population movement in response to climate-related hazards in Bangladesh Environmental Science & Policy (2013), pp. 44-59. DOI: 10.1016/j.envsci.2012.03.009 Perry, R.W. (1979). Evacuation decision-making in natural disasters. Mass Emergencies 4: 25–38. Rashid, H. (1991). Geography of Bangladesh. University Press Ltd., Dhaka. Rashid, H. and 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. Sultana, P, Johnson, C. and 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. Thompson, P.M. and P. Sultana. (2000). "Shortcomings of Flood Embankment Strategies in Bangladesh." In Parker, D.J. (ed.) Floods (vol. 1). (London: Routledge) pp 316-333. Thompson, P. and Tod, I. (1998). Mitigating flood losses in the active floodplains of Bangladesh. Disaster Prevention and Management 7: 113–123. World Bank. (1990). Flood Control in Bangladesh: a Plan for Action. (Washington: Asia Region Technical Department, World Bank).

Hydro-Social Deltas Migration and hazards 2016 - Annex 1 A1-1

ANNEX 1 - DETAILED RECORDS OF FOCUS GROUP DISCUSSIONS

1.1 Lalmonirhat

Mainland Satata Rajapur Federation Progoti Tushvander Federation Onnesha Velabari Federation Location Union: Rajpur

Upazila: Sadar District: Lalmonirhat

Union: Tushvander Upazila: Kaliganj District: Lalmonirhat

Union: Velabari Upazila: Aditmari District: Lalmonirhat

Number of villages covered by CBO

The Federation includes 22 villages, out of which 60% of Rajapur village and 50% of Thakurdari village was eroded , 30% of 7 villages was eroded, 75% of two villages are in char.

9 villages (Tushvander, Sundrahobi, Talikbarinagar, Kanchanshar, Dakshin Ghanasham, Uttar Ghanasham, 2 are eroded (Kashiram-1 and Bairati-3) few years back but now again accreted and occupied. Talikbarinagar is completely save by embankment

10 villages Falimari (part char-adajacent)-1, Falimari-2, Velabari(part char-adajacent), Shalmara (part char-adajacent), Mahishtali (part char-adajacent), Talukdulali, Paglarchara, Pashchim Velabari, Purba Velabari, and Mohammmadpur.

Land level 70% high, 30% low 66% high, 33% low 80% high land, 20% low

Adjacent River Tista Tista Ratnai Nadi connected to Dharla Number of households within the site

6000 9000 5256

Area 2560ha 3650ac 7941ac Protected area 80% Tushvander, Sundrahobi, Talikbarinagar 70% of

Dakshin Ghanasham, Uttar Ghanasham 30% of Bairati, 40% of Kashiram-1,Kashiram-2 are protected by embankment. In Kanchanshar 100% of the embankment is eroded.

56%

Unprotected 20% 30% 44% Occupation Most of the people Re involved in agriculture,

petty trade in the town and work as wage labour.

30% people are fishers, 20% cultivate their own land and rest work as labour outside the area.

50% migrate seasonally for work, 35% work as agricultural labour and rest 15% live on fishing

Land eroded in 2016 20% land eroded in 2016. Villages in chars suffered erosion more than the villages in mainland. Erosion happens every year but this year the homesteads eroded more than the agricultural land.

100% land of one big village eroded, Part of two newly formed chars eroded. 200 homesteads eroded and 100 households moved temporarily on the embankment and 100 households moved to the protected area


Land flooded in 2016 60% (10 villages and chars) 40% 25% Frequency of flood in 2016 According to respondents flooding was less

than in 2015. In 2015 the flood water stayed on the land for about two weeks. On the charland the water level was about 6 feet and on the mainland it was 4ft.. This year flood hit 3 times (2 times in July and once in August)

Erosion was high this year and flood hit the area due to 4 times, 2 times in July/August and stayed for 15 days and once in August and once in September and water stayed for 2 weeks. The one in August was worse than the others. Erosion was high this year than other years.

This year flood came about 4 times-2 big ones in July/August Which stayed for 10 days each time and two small ones in August/September which stayed for 7days.


Mainland Satata Rajapur Federation Progoti Tushvander Federation Onnesha Velabari Federation and stayed for 5 days. Water level on the agricultural land was about 2.5ft and on the homestead water level was about 1 ft.

Frequency and extent of erosion are higher this year than in last 10 years. Homesteads were under 4 to 4.5ft high water.

Households eroded 800 200 0, The homesteads are far from the river bank. Households flooded 1200 400 100 Duration of standing flood water

5-7 days depending on land level 10 to 15 days depending on land level 7-15 days depending on land level

Height of water Homestead: 1 ft Agric land: 2-2.5ft

Homestead: 4-4.5 ft Agric land: -6-7ft

Homestead: 1 ft Agric land: 5ft

% of households evacuated due to flood

460 400 households 0

Migration 10hh moved to mainland. 50households moved to Paglarhat village. About 300 households moved to India. Some moved to Rajarhat and Rajibpur upazila and others moved temporarily to adjacent areas. Poor try to stick to the same area.

All 400 households flooded in homestead moved within in the protected area, on the embankment, higher ground and shelters, school ground.

People managed to stay at home and did not move

Coping 75% household male members seasonally migrate to Dhaka and Chittagong (construction work, brick field, pulling rickshaw, professional work like carpentry, Masonry) Comilla, Noakhali, Bogra and (agricultural labour) and Mushiganj for planting and harvesting potato. For coping they also sold land, took loan from NGO and from Federation. They also work overtime to earn more money.

60% household male members from unprotected area and 40% from the protected area migrate to Dhaka, and Chittagong (construction work, brick field, pulling rickshaw,) Comilla, Tangail and Bogra (for agricultural labour). 20% of the households catch fish in monsoon season and only 10% area involved in risky agriculture.

50% household members from unprotected area move temporarily every year for 6-8 months in search of work. Out of which 15% go to Dhaka and adjacent area to work in the garment industry. Other go seasonally to Bogra and Sherpur area to plant and harvest Aman paddy.

Duration of temporary/seasonal migration each time

Depends on types of work. They usually come home every 3 to 4 months but usually stay out of their home for 6 months. Those who has agricultural land they come home frequency to help family. Those who goes to work as agricultural labour they go on a contract and in group. Usually they go to the same area they are familiar with. In last 20years 5% households moved 2 to 5 times but never moved more than 2 miles. Some participants mentioned moving 8-10 times since 1990 but never moved more than 2 miles. However some people moved 2 to 3 times within last one year.

On an average people stay away from home for 4 to 6 months. The labour class work for 30 to 60 days, come home and again go. Those who go for agricultural work go twice, once for sowing/transplanting and once for harvesting. Those who work in garment industry or other industry/factories go for longer time. They usually come for holidays.

Those who goes for agricultural work and wage labour (construction labour, brick field etc.) usually work for 4-6 months a year. Those who pulls rickshaw work for for 15-21 days and come home for a week or two and then they again go to cities.

Pros and Cons of migration During flood people temporarily go to the About half of people moved to embankment There was no evacuation. People lived in their


Mainland Satata Rajapur Federation Progoti Tushvander Federation Onnesha Velabari Federation higher ground, schools, embankment. Respondents said that the place become overcrowded, lack of drinking water and latrine and place for cooking as none provide cooked food. Relief was not enough. Women suffer from lack of security and privacy. Males from almost every household migrate seasonally/temporarily outside the area for work. As there is very few opportunities for work in the area, the male members have to go to the cities and even to other places for work. They earn money there to maintain their families. Those who earn a better wage can afford to rent a room where 4/5 people share. Those who are construction worker often sleep in the site. The agricultural labourers have to find out places to sleep nearby their contraction sites. They go in a group and stay there until their specific job is finished. But getting work everyday is uncertain for other wage labour. In the dry season work is available but in monsoon they have to come back home. They have to work hard to earn money. Moreover, they are always in worry about the family they left at home. Sometimes people fall in trap and get involve in drugs and other crime. All of them complained about quality of accommodation and food they eat. Construction labourers complained about getting their wages on time and they never get full payment. Those who pull rickshaw/van sleep on the road in shacks made of plastic paper. All of them said that they face water, sanitation, food and shelter problem. Besides, rickshaw puller have to pay police. Others like street vendors have to pay local mastans and police. Women living at home take care of the children and the family, livestock etc. They have to face different problems (health, food etc.) all by themselves. Sometime they do not get money from their male partners on time.

temporarily. During temporary evacuation on the embankment or in the shelter they suffer from space problem. Too many people have to stay in the same place crowded where cooking and sleeping, sanitation are bigger problem. Getting clean drinking water most of the time causes water borne diseases. Women face problems like unhygienic environment and lack of security. People who leave their original place tend to stay in unhygienic environment and insecure places. Men who migrate to cities for work usually sleep in crowded places, mosques, park benches, footpath railway station etc. They have to face problem for food, clean drinking water, toilet and also theft. Those who come with family often end up sleeping on roadside shacks which they have to dismantle during day time. Shortage of drinking water, toilets are expose them to health problem. However, they also mentioned about travel cost to and from the working place. Women left in the house have to bear all the problems at home-shortage of money (as men may not be able to send money on time), health problem of herself and the children etc. The good thing is that they can earn there and can send money for the maintenance of the family left behind. Besides, they have to borrow money in advance to give to the family before they leave at a high interest rate. Some people get married again and never come back to the village. Some family send their school going children for work to cope with the situation.

own house. When men migrate to cities or to other places for work they have to live in unhealthy environment facing problem of cooking, accommodation and sanitation. They are forced to sleep in mosques, railway station, bus station, under the shade of shops in local markets etc. If the contract for agricultural labour includes food they get food but mostly for other work they have to cook themselves. For the other labours this is also applicable. Rickshaw pullers have to do hard work and they cannot continuously work for days. Often they fell sick and weak. But they earn for the family. If they stay back they cannot earn anything in their own locality. Some fell in the trap of Mahajan who give loan on high interest rate and rip off all the money they earn. The federation have their own credit system and they can borrow from the federation. Some get advance for their work and has to fulfill their job at any cost. In case of garment workers, the main problem is job security and personal security. These people get less salary than the others. There is problem for getting rental houses. The ones they get can not afford. They are forced to stay in low cost houses and in crowded places where they are not connected to water supply and gas.


Mainland Satata Rajapur Federation Progoti Tushvander Federation Onnesha Velabari Federation They have to loan from local Mahajans on high interest.

Permanent migration People who have money they buy land outside the area and go permanently, but the percentages are very small. Some went to India and never came back. People go to cities, specially women for work in garment industry can only come home during festivals or long holidays. They only get one day off weekly. But some work overtime to make more money.

This year very few people migrated permanently as they wanted to stick together with their family and be secure and get help from each other. They said they are afraid of moving from their birth place and they do not have money to buy better land outside the area and they do not know other there.

None migrated permanently from this area. But from 20% household members go to cities to work in the garment industries.

Duration stayed outside the homestead in 2016 after flood

15 days 15 days

Status of livestock during flood/erosion

Cattle were moved to higher ground such as school ground, embankment and higher land and remained there for 15 days until the homestead dries up. Very few livestock died due to feed and proper shelter. Too many cattle in the same space spread disease. People usually keep cattle as safety net. In 2016 they fattened cattle before flood and sold them during Muslim festival Eid-ul-Azha. But they could not get proper price as middlemen paid less knowing their inability to sell cattle by themselves due to flood and shelter. This is a common practice. As India restricted cattle movement from India to Bangladesh and there was huge demand for cattle during this festival, people fattened cows to sell.

Cattle were moved to the protected area where there is higher ground, near school premises and on embankment and remained there for a month till they feel secured. About 1% livestock died and 10% were sold to recover from the after flood shock. Usually it is not easy to maintain livestock in the mainland as most of the land are cultivated most of the year. They also take their livestock to chars during dry season for grazing. During flood they have to stall feed cattle but the feed they can provide was not enough and the livestock health was at stake. But they can sell livestock during at high prices

About 80% households keep livestock. When flood comes they move the livestock in safe places. 60% households sell their livestock during monsoon or before monsoon.

Status of embankment Embankment was breached in 3 to 4 places. Although there are 3 groins due to heavy current in 2016 embankment breached.

Eight km embankment was not breached and people living inside the embankment was safe.

The embankment was built in 1989. There was several breaches different as they river current was very strong. But government has repaired it. There is a rubber dam to keep the water level controlled and can preserve water for fish culture and irrigation.

Initiatives to help people Early warning was given by different NGOs, and Government and Union Parisad. However, people saw the river current and realized that there will be mass disaster. They moved to higher ground in the main stable area. CBO has helped them to move.

Early warning by NGO and Government was provided. Volunteers were trained before on preparedness and raise awareness among people. There is a Village Disaster Management Committee (VDMC) in the area. Besides hoisting flag, miking was also performed by the

In 2016 there was no erosion. New chars accreted. The CBO members did not get receive much help as there was very little damage. They got help from the federation.


Mainland Satata Rajapur Federation Progoti Tushvander Federation Onnesha Velabari Federation volunteers. Embankment and schools were used as shelter. Government provided some relief although people complained it to be not enough.

CBO role during flood/erosion One of the main activity of the CBO is disaster management. The CBO members received training on early management and formed volunteer group. They relocate people by using their boats and man power, find out space for the people and their livestock to move. They report to RDRS (a regional) NGO for relief and help. They also keep contact with the government departments for post flood/erosion rehabilitation. They provide early warning through own miking system. They also cooperate Union Parisad during relief distribution.

CBO members have helped people to move, provided some help for a day. They prepared list of flood and erosion victims, appealed o NGO and government agencies for help for the victims. They also communicate with the Union Parisad. They formed rescue team and helped people to move to safe places with boat. They also Help people to rebuild their houses and repair roads. They also provided information on the flood and erosion to their members. Distributed seeds, tubewell, latrine and medicine among the erosion and flood victims after flood and erosion.

The CBO has a Disaster Management Committee. They have provided information on the flood and erosion to their members. In 2016 they provided Early warning and helping vulnerable households to the safer places before flood. They also provided information on affected people of the area and distributed relief provided by the government and INGOs.

Impact of embankment Embankment specially groins save the villages from massive erosion this year. However, A char is forming and the residents hope to have less erosion next year. Embankment is used as shelter during flood and erosion and also used as road for communication. Respondents said it was “inhuman” to live on the embankment as it was so crowded with char and flooded mainland people. Embankment is also used as road during dry season. There is a railway track which is also providing protection.

Embankment have saved people and providing shelter to the flood and erosion victims. Due to embankment sand deposition on the agricultural land inside the protected area was avoided. However, water logging from torrential rain also impacted people’s life inside the protected area . Erosion victims build houses and live on the embankment.

Embankment breached but people take action right away to protect the embankment. Embankment works as shelter for the flood and erosion victims.

Disasters in last 10 years Flooding happens every year but was high in 2015 and in 2011. Erosion every year but severe in 2015 and 2016 Drought: 2011 and 2013 Excessive rainfall: 2011, 2016 Now a days other natural calamities decreased.

Flood causes misery every year but in little more in 2004, 2008, 2011 and 2014. Erosion every year but severe in 2004, 2008, 2015 and 2016. Char formation is faster now than before. People do not have to wait for 30-40 years to get their land back. Drought: 2003, 2014 and 2015 Excessive rainfall: 2001, 2005, 2006 and 2014 Other natural calamities decreased but thunderstorm and fog increased now.

In 2016 Ratnai river current was stronger than in 10 years back, erosion and accretion are frequent. It takes 2-3 years to accrete land now in comparison to 30-40 years before. According to the respondents, flood was severe in 2015 than in 2016. But they suffer from heavy rain and strong wind in 2016. Flood also affected the area every year but it was little higher in 2011, 2012, 2013 and in 2014 than other year. According to the respondents water level in the area was higher on the homestead and in the agricultural field in 2015. Drought affected the area in 2014,


Mainland Satata Rajapur Federation Progoti Tushvander Federation Onnesha Velabari Federation 2015 and in 2016. Water stands on the ground for more than a month in 2015 and the water level was 1ft high than 2016. Thunder storm killed several people and livestock in this area in 2015. Fog in 2012 and in 2015 caused severe damage to the winter crops.


Damaged houses, infrastructure, crops (Jute and Paddy), seedbeds were under water. In exposed areas

Some people have to move several times. They have to dismantle their shacks and build it again. They lost all their agricultural land, they have to sell all their livestock.

Impact of erosion Houses eroded, agricultural land eroded, people moved to safer places.

Houses eroded, agricultural land eroded. As they have to stay in the open space they are exposed to rain and thunder shower. They have to move which caused financial loss. They have to sell some livestock and be separated from the original community.

Very few houses eroded but lost standing crops.

Coping with Flooding Loan from NGO, Mahajan, sell livestock, agricultural land, migrate for work for longer time

Loan from NGO, Mahajan, sell livestock, catch fish in the river, migrate to work in other cities.

Loan from Federation and Mahajan, sold livestock, sold fixed assets, lease land out, migrate to work in other cities.

Coping with Erosion Migration, Loan from NGO, Mahajan, sell livestock

Loan from NGO, Mahajan, sell livestock, catch fish in the river, migrate to work in other cities

Loan from Federation and Mahajan, sold livestock, catch fish in the river, migrate for work in other cities.


1.2 Gaibandha

River side Konai Brahmaputra Community Based Fisheries Organisation

Asher Alo Unnayan Sangstha(1) Asher Alo Unnayan Sangstha (2): Vati Bochagari Unnayan Samity

Location Union: Gazaria Upazila: Fulchari District: Gaibandha

Union: Kamarjani Upazila: Sadar District: Gaibandha

Union: Kapasia Upazila: Sundarganj District: Gaibandha


5 villages Khamarpara (char, eroded), Baluchar, Gobindi, Nilkuthi and Katlamari)

4 villages Karaibari(char) and Goghat 1,2,3(facing erosion)

4 villages Ujan Bhurail (char), Vati Bhurail (char) Rajarchar (char) and vati bochagari

Land level 60% high, 40% low 30% high, 70% low In Ujan Bhurail village 40% land exist, in Vati Bhurail village 17% eroded, in Rajarchar village 20% eroded and vati bochagari 100%

Adjacent River Brahmaputra ( Jamuna) Brahmaputra Brahmaputra and Teesta Number of households within the site

500 800 3000

Area 20sq.km 1175ha 780ha Protected area 20% 0% 0% Unprotected 80% 100% 100% Occupation Most of the people go fishing, do agriculture

on leased land, share crop or work as labour. Some work in the town as day labourers.

40% people are fishers, 10% cultivate their own land and rest work as labourers, go outside the area for work.

75% migrate seasonally for work, 5% work as agricultural labourers and rest 20% live on fishing

Land eroded in 2016 35%, every year erosion happens but the year when current is high there is more erosion happened.

80% land of one big village eroded, all homesteads eroded and the 300 households moved temporarily and 100 households moved permanently from the area.

52%. 100% people live on Khas (government) land.

Land flooded in 2016 50% 100% 100% Frequency of flood in 2016 According to respondents this is a big flood

after 1988 one. There was flood in 1998 but it did not stay for longer. The flood in 2016 caused more damage than 1988 flood. Flood came about 5 times-once in early June which stayed for 5/6 days. The second one came in July and stayed for 15 days. The third one came in August and stayed for 1 month. This was the devastating one as it causes damage to standing crop. The fourth one came in September and stayed for 18/19 days and the last one came in end September and stayed for 5/6 days. Farmers lost second time planted crop in end September one.

Flood in this area was one feet lower than flood level in 1988. But this year erosion and food hit at the same time making life difficult as they have not been able to come back home within a month. Flood hit hard the area thrice once in July/August and stayed for 20 days. Erosion was high this year than other years. Frequency and extent of erosion are higher this year than in last 10 years. Another two in September which stayed for 10-12 days. Homesteads were under 5 to 6 feet water. Even the highlands went under 1-2 feet water.

This year flood came about 4 times-2 big ones in July/August Which stayed for 20 days each time and two small ones in August/September which stayed for 10 days. Erosion was high this year than other years, but extent increased from last year.

Households eroded 230 600 750




Households flooded 250 400 2400 Duration of standing flood water

10 to 20 days depending on land level 15 to 30 days depending on land level 20 days

Height of water Homestead: 3-5 ft Agric land: 6-10ft




50 75 80

Migration 90hh moved outside the area (Fakirhat, Kalirbazar, Bonarpara, Taluk Kanupur, Nakai hat), but not too far. 100 moved to embankment and flood shelters. Others moved temporarily to adjacent areas. Poor try to stick to the same area.

250 households moved within CBO area to embankment, higher ground and shelters, school ground. 50 households move to adjacent area specially to adjacent area Sundarganj, Malibari, Dariapur. These people will move back when new char will be visible.

2400 hh moved to nearby villages, within protected area, on the embankment and on higher ground. Rest stayed back in their original places, living on macha (raised platform). Some have raised house where they and their livestock live.

Coping 25-30% household male members Seasonally migration to Dhaka, Chittagong, Sylhet, Munshiganj and Cox’sbazar

70% household male members migrate to Dinajpur, Thakurgaon, Rangpur, Dhaka, Munshiganj and work as wage labour in brick field, Garments, drive rickshaw/ van/ autorickshaw.a big percentages of the male and some female also go to Munshiganj for planting and harvesting potato. 40% of the households catch fish in monsoon season and only 10% area involved in risky agriculture.

75% household members move temporarily every year for atleast 6 months in search of work. During dry season 10% people go to adjacent area (Sreepur, Chadipur) for harvesting Jute and maize. Most of the people migrate to Dhaka and Chittagong to work as labourers in brick field, pull rickshaw and work as construction labourers. They also go to Manikganj and Bogra for harvesting rice. Migration to Munshiganj is for planting and harvesting potato. Rest 6 months they mostly sit idle. They usually sell livestock before monsoon as they do not have permanent place to live and security for the livestock. They also do sharecropping.


Depends on types of work. On an average most of the people go for 6 months (between October and June) in a year. In cities like Dhaka and Chittagong they work as rickshaw/van puller, earth cutting labour, construction labour, stone carrier etc. In Mushiganj area they work as agricultural labourers specially for planting and harvesting potato. In Sylhet area they go for rice harvesting. Some people work in factories and live in Gazipur, Joydebpur, Kalagachi and Mirpur. In almost all cases they go in

On an average people stay away from home for 6 to 8 months. The fishers fish in group in the river and stay away for 6 months a year. The labourers work for 30 to 60 days, come home and again go. Those who go for agricultural work go twice, once for sowing/transplanting and once for harvesting. They go to Dinajpur, Bogra and Rangpur area. Those who work in garment industry or other industry/factories go for longer time. They usually come for holidays.

Due to lack of work and extreme risk, people spend more time outside the area for work. They spend at least 8 months during dry period and come back to their original place during monsoon to help their family move or cope. They do any work they can get for survival.




group to find job. Pros and Cons of migration During temporary evacuation, respondents

said that the place become overcrowded, lack of drinking water and latrine and place for cooking as none provide cooked food. Males from almost every household migrate seasonally/temporarily outside the area for work. They earn money there to maintain their families. Those who earn a better wage can afford to rent a room where 4/5 people share. Those who are construction worker often sleep in the site. The agricultural labour have to find out places to sleep nearby their contraction sites. They go in a group and stay there until their specific job is finished. But getting work everyday is uncertain. They have to work hard to earn money. Moreover, they are always in worry about the family they left at home. Sometimes people fall in trap and get involve in drugs and other crime. All of them complained about accommodation and food they eat. Construction labourers complained about getting their wages on time and they never get full payment. Those who pull rickshaw/van sleep on the road in shacks made of plastic. All of them said that they face water, sanitary, food and shelter problem. Besides they have to pay local mastans and police. Women living at home take care of the children and the family, livestock etc. They have to face different problems (health, food etc.) all by themselves. Sometime they do not get money from their male partners on time. They have to loan from local Mahajans on high interest.

During temporary evacuation in the shelter they suffer from space problem. Too many people have to stay in the same place crowded where cooking and sleeping, sanitation are bigger problem. Getting clean drinking water most of the time causes water borne disease. Women face problems like unhygienic environment and lack of security. People who leave their original place tend to stay in unhygienic environment and insecure places. Men usually sleep in crowded places, mosques, park benches, footpath, railway station etc. They have to face problem for food, clean drinking water, toilet and theft. Men who goes alone sometimes get married and never comes back. Those who come with family often end up sleeping on roadside shacks which they have to dismantle during day time. Shortage of drinking water, toilets are expose them to health problem. Women left in the house have to bear all the problems at home-shortage of money (as men may not be able to send money on time), health problem of herself and the children etc.

During temporary evacuation majority moved to embankment, some made shacks on the higher ground near school, Madrasa, and killa. Migrants have to live in very crowded places, some people even do not get space for sleeping, due to mosquito and lack of clean drinking water they fell sick. For women latrine is a problem. According to the participants some NGO helped people to raise plinth where they can stay. Some people live on the boat. Cooking food is problem for dry fuel and space. They cook one in a day and eat that. Getting work is a problem. However, they mentioned relief is a positive point for them although it is not enough. When men migrate to cities or to other places for work they have to live in inhuman environment having problem of food, shelter and water. They are forced to sleep in mosques, railway station, bus station, under the shade of shops in local markets etc. If the contract for agricultural labour includes food they get food but mostly for other work they have to cook themselves. For the other labourers this is also applicable. Rickshaw pullers have to do hard work and they cannot continuously work for days. Often they fell sick and weak. But they earn for the family. If they stay back they cannot earn anything in the locality. Some fell in the trap of Mahajan who give loan on high interest rate and rip off all the money they earn. No NGO give loan in this area s they are floating people and has no permanent address or property. Some get advance for their work and has to fulfill their job at any cost.

Permanent migration People who have money they buy land outside the area and go permanently, but the percentages are very small. About 30 households migrated to Dinajpur and never

Very few people migrated permanently as they wanted to stick with their extended family and be secure and get help from each other. They said they are afraid of moving from their birth place

None migrated permanently as they get supports from the neighbor and from NGO.




came back. and they do not have money to buy better land outside the area.


1 month 20 days 1 month


Cattle were moved to higher ground such as school ground, embankment and higher land and remained there for one month. About 30% of the livestock died due to feed and proper shelter. Too many cattle in the same space spread disease. People usually keep cattle and fatten for selling. This is a common practice. Specially in 2016 people raised cattle with the hope to make profit during Eid-ul-Azha (sacrificing livestock). As India restricted cattle movement from India to Bangladesh and there was huge demand for cattle during this festival, people fattened cows to sell.

Cattle were moved to higher ground such as school ground, embankment and higher land and remained there for 20 days. About 2% livestock died and 5% were sold to recover from the after flood shock. Usually it is easy to maintain livestock in the chars and people like to raise cattle. During flood they have to stall feed cattle but the feed they can provide was not enough and the livestock health was at stake. But they can sell livestock during Muslim festival.

They do not keep any livestock as they have to move several times in a year.

Status of embankment behind the villages

About 0.5 km of the embankment breached between Katlamari and Singria due to high current pressure in the river.

Embankment was not breached and people inside the embankment was safe.

Initiatives to help people Early warning was given by different NGOs, Government Information Services, Red Crescent , Social welfare department and Union Parisad. However, schools were used as shelter and people moved by themselves as everyone was affected and very difficult to help each other.

Early warning by NGO and Government was provided. Volunteers were trained before on preparedness and 40 (20 from young group and 20 from community) received training from NGO. Besides hoisting flag, miking was also performed by the volunteers. Embankment and schools were used as shelter.

In 2016 two villages Vati Bochagari and Rajar char was eroded three times. Since 1988 char started to erode and since then upto 2015 these chars eroded 7 times. People moved to the land area with the help of the NGO. NGO provided information on flood and erosion. Practical Action, an INGO, raised 100 house plinth and renovated house, built latrine and provided tubewell. In 2007 erosion victims who moved to embankment received cattle.

Government and NGOs provided dry food for one day only.

Government provided relief (rice, dal salt). Government provided relief (rice, dal salt).

CBO role during flood/erosion CBO members are also very poor. Most of them fish for living. They helped people to move and when needed they helped them to build other’s shack. Members collect bamboo from whoever have Bamboo groves for house rebuilding on the embankments. They also provided early warning messages to their members and their trained volunteers gave

CBO members have helped people to move, provided some help for a day. They provided dry food, formed rescue team and helped people to move to safe places with boat. They also provided information on the flood and erosion to their members. Distributed seeds among the erosion and flood victim.

CBO provided information on the flood and erosion to their members. They also distributed relief provided by the government and INGOs.




training on how to move to a safer place, how to preserve food and essentials etc.

Impact of embankment As these people were living outside the embankment they took shelter on the embankment. A big breach in embankment make the people living inside the embankment vulnerable to flooding. Respondents said it was “inhuman” to live there as it was so crowded with char and flooded mainland people. Embankment is also used as road during dry season.

Embankment have saved people providing shelter on it. Due to embankment sand deposition on the agricultural land inside the protected area was avoided. However, flood from torrential rain also impacted people’s life within the protected area. Erosion victims build houses and live on the embankment.

Embankment breached in 20 places during the flood in several places. From inside due to heavy rain water logging and pressure from outside current causes the breach. Local people used sand bags and put earth in the breached area to protect the embankment. Embankment works as shelter for the flood and erosion victims.

Disasters in last 10 years Flooding in every year but bit more in 2004, 2005,2008, 2011 and 2014. Erosion every year but severe in 2015 and 2016 Drought: 2011 and 2013 Excessive rainfall: 2011, 2016 Now a days natural calamities decreased.

Flood causes misery every year but in little more in 2004, 2005,2008, 2011 and 2014. Erosion every year but severe in 2015 and 2016 Char formation is faster now than before. People do not have to wait for 30-40 years to get their land back. Drought: 2011 and 2013 Excessive rainfall: 2011, 2016 Other natural calamities decreased but thunderstorm increased now.

River current was stronger in 2016 than 10 years back, river bed is silted, erosion and accretion are frequent. It takes 2-3 years to accrete land now in comparison to 30-40 years before. Heavy rainfall and strong wind in 2016 and 2013, Drought in 2014, Thunder storm killed 16 people in this area in 2015-16.


Damaged houses, infrastructure, crops (Jute and Paddy), seedbeds were under water. Lots of sand deposition on the agricultural land. Farmers are afraid that may be in next two years they will not be able to harvest good crop

Some people have to move several times. They have to dismantle their shacks and build it again. They lost all their agricultural land, they have to sell all their livestock.


Houses eroded, agricultural land eroded. As they have to stay in the open space they are exposed to dacoity.

Houses eroded, lost standing crops.

Coping with Flooding Loan from NGO, Mahajan, sell livestock, fish business


Loan from Mahajan, sold livestock, catch fish in the river, migrate to work in other cities.



Loan from Mahajan, sold livestock, catch fish in the river, migrate to work in other cities.


1.3 Kurigram

Chars Jatrapur federation Goruhara Gram Unayan Samity Porarchar Gram Unnayan Samity Location Union: Jatrapur

Upazila: Sadar District: Kurigram

Union: Panchgachia Upazila: Sadar District: Kurigram

Union: Begumganj Upazila: Ulipur District: Kurigram


The Federation includes 18 villages, out of which 6 villages are river char and 50% of other villages were partly in char.

9 villages (Kadamtala, Sitaijhar, Balamajhir char, Dakshin Goruhara, Kamarchar, Goldarchar, Dharalar char, Mazhipara, Dhakshin Nowabash)

7 villages(Mondolpara, Hazipara, Kalimuddinpara, Memberpara, Porarchar, Sluice gate para and Inol Haque Para).

Land level 65% high, 35% low 50% high, 50% low 60% high land, 40% low

Adjacent River Dharla and Brahmaputra Dharla (west)and Brahmaputra (east) Brahmaputra Number of households within the site

5825 2000 960

Area 70sq.km 1800ac 941ac Protected area 33% 0% 0% Unprotected 66% 100% 100% Occupation 60% of the people are involved in agriculture,

rest in petty trade in the town and work as wage labour.

20% people are fishers, 30% cultivate their own land and rest work as labour outside the area.

70% migrate seasonally for work, 20% work as agricultural labour and rest 10% live on fishing

Land eroded in 2016 25% land eroded in 2016. Villages in chars (Garuhara-50%, Parbatipur-25%, char Jatrapur mouza) suffered erosion more than the villages adjacent to mainland. Erosion happens every year but this year the agricultural land eroded more than the homestead. The water current was higher than other years.

100% land of Kadamtala and Dakshin Goruhara village eroded, in other char villages about 15-20 households eroded partly.


Land flooded in 2016 66% (12 villages and chars) 100% 100% Frequency of flood in 2016 According to respondents flood height was 2

ft less than in 1988. In 2016 the flood water stayed on the land for about 16 days. On the unprotected area the water level was about 10-12 feet and on the protected area it was 5ft. On the homestead the water was as high as 5-7ft This year flood hit 3 times ( once in July and once in August and once in September). The big one in July stayed for 16 days but the August/September one stayed for 5 days.

Flood hit the area 3 times, 2 times in early July and early August and stayed for 20 days and once in August and water stayed for 2 weeks. The one in early July was worse than the others

This year flood came about 3 times-one big one in July/August Which stayed for 15-20 days each time and two small ones in August/September which stayed for 7days. Water stayed for one month on the agricultural land.

Households eroded 27, Within 10km land area in last 10 years only 1.5 km area exists. Now the people living in that area has no land. Before it took 15 to

In two villages all households eroded and in other chars 15-20 households eroded .

20 The homesteads are far from the river bank.


Chars Jatrapur federation Goruhara Gram Unayan Samity Porarchar Gram Unnayan Samity 20 years to accrete land and char arose far from the eroded area which influenced violent conflict. Now chars arose faster, within 5/6 years and near the eroded char. So it is easier to locate their own land. The reason they mentioned is depth of river and braiding of river.

Households flooded <2000 2000 900 Duration of standing flood water

5-16 days 15 to 30 days depending on land level 15-20 days in the homestead and one month on the agricultural land.

Height of water Homestead: 4-5 ft Agric land: 5-12ft depending on land level

Homestead: 4-5 ft Agric land: -10-15ft

Homestead: 3-4 ft Agric land: 7-8ft Lowlands: 12ft


1700 200 households 14

Migration 1000 moved to mainland, on the ambankment, higher ground and to the adjacent mainland areas. Others make high platform and lived there. Some lived on the roof, some lived on the boat.

All 400 households moved within in the protected area, on the embankment, higher ground and shelters, school ground.

Most people managed to stay at home on raised platform and boats and did not move

Coping 60-70% household male members seasonally migrate to Dhaka and Chittagong (garments, construction work, brick field, pulling rickshaw, professional work like carpentry, Masonry) Comilla and Noakhali, (agricultural labour). For coping they also sold land, livestock, left the area in search of work, took loan from NGO and from Federation. They also work overtime to earn more money.

75% household male members migrate to Dhaka(construction work, brick field, pulling rickshaw,) Comilla, Feni , Munshiganj and Narayanganj (for agricultural labour). 20% of the households catch fish in monsoon season and but 75% are involved in risky agriculture as sand deposited on the crop land every year making land unfertile.

75% household members from unprotected area move temporarily every year for 6-9 months in search of work. Out of which 10% go to Dhaka and adjacent area to work in the garment industry. Other go seasonally to Dhaka, Chittagong (construction work, brick field, pulling rickshaw,) Comilla, Feni , Munshiganj area to plant and harvest Aman paddy and Potato.


Depends on types of work. They usually come home every 2 to 3 months but usually stay out of their home for 6 months. Those who has agricultural land they come home frequency to help family. Those who go to work as agricultural labourers they go on a contract and in group. Usually they go to the same area every year as they are familiar with the land owners and the area. In last 10years 5% households moved 2 to 5 times but remained within the adjacent area. Some participants mentioned moving 10-12 times

On an average people stay away from home for 6 to 8months. The labourers work for 30 to 60 days, come home and again go. Those who go for agricultural work go twice, once for sowing/transplanting and once for harvesting. Those who work in garment industry or other industry/factories go for longer time. They usually come for holidays.

Those who goes for agricultural work and wage labour (construction labour, brick field etc.) usually work for 4-6 months a year. Those who pulls rickshaw work for 15-21 days and come home for a week or two and then they again go to cities.


Chars Jatrapur federation Goruhara Gram Unayan Samity Porarchar Gram Unnayan Samity since 1990 but never moved more than 2 miles. However some people moved 2 to 3 times within last one year. 10 years back less people were migrating and going to adjacent towns or areas such as Rangpur, Bogra etc. But now more people move and go to further places where they can get work.

Pros and Cons of migration During flood people temporarily go to the higher ground, schools, embankment, roads. Some people stick in their homestead or on the boat near their homestead. Respondents said that the shelter place become overcrowded. They face shortage of drinking water and latrine and place for cooking as none provide cooked food. Relief was not enough. Women suffer from lack of security and privacy. Males from almost every household migrate seasonally/temporarily outside the area for work. As there is very few opportunities for work in the area, the male members have to go to the cities and even to other places for work. They earn money there to maintain their families. Those who earn a better wage can afford to rent a room where 4/5 people share. Those who are construction worker often sleep in the site. The agricultural labour have to find out places to sleep nearby their contraction sites. They go in a group and stay there until their specific job is finished. But getting work everyday is uncertain for other wage labour. In the dry season work is available in the area but in monsoon they have to come back home to help their families. They have to work hard to earn money. Moreover, they are always in worry about the family they left at home. Sometimes people fall in trap and get involve in drugs and other crime. All of them complained about quality of accommodation and food they eat. Construction labourers complained about getting their wages on time and they

About half of people moved to embankment and protected mainland temporarily. During temporary evacuation on the embankment or in the shelter they suffer from space problem. Too many people have to stay in the same place crowded where cooking and sleeping, sanitation are bigger problem. Getting clean drinking water most of the time causes water borne diseases. Women face problems like unhygienic environment and lack of security. People who leave their original place tend to stay in unhygienic environment and insecure places. Men who migrate to cities for work usually sleep in crowded places, mosques, park benches, footpath railway station etc. They have to face problem for food, clean drinking water, toilet and also theft. Those who come with family often end up sleeping on roadside shacks which they have to dismantle during day time. Shortage of drinking water, toilets are expose them to health problem. However, they also mentioned about travel cost to and from the working place. Women left in the house have to bear all the problems at home-shortage of money (as men may not be able to send money on time), health problem of herself and the children etc. The good thing is that they can earn there and can send money for the maintenance of the family left behind. Besides, they have to borrow money in advance to give to the family before they leave at a high interest rate. Some family send their school going children for work to cope with the situation.

There was very little evacuation. People lived in their own house due to fear of theft and with the hope that water will recede quickly. When men migrate to cities or to other places for work they have to live in unhealthy environment facing problem of cooking, accommodation and sanitation. They are forced to sleep in public places, on the sidewalks, mosques, railway station, bus station, under the shade of shops in local markets etc. If the contract for agricultural labour includes food they get food but mostly for other work they have to cook themselves. For the other labourers this is also applicable. Rickshaw pullers have to do hard work and they cannot continuously work for days. Often they fell sick and weak. But they earn for the family. If they stay back they cannot earn anything in their own locality. Some fell in the trap of Mahajan who give loan on high interest rate and rip off all the money they earn. The CBO has it’s own credit system and they can borrow from the CBO. Besides NGOs also provide short term loan. Most takes advance for their work and has to fulfill their job at any cost. In case of garment workers, the main problem is job security and personal security. These people get less salary than the others. There is problem for getting rental houses. The ones they get can not afford. They are forced to stay in low cost houses and in crowded places where they are not connected to water supply and gas.


Chars Jatrapur federation Goruhara Gram Unayan Samity Porarchar Gram Unnayan Samity never get full payment. Those who pull rickshaw/van sleep on the sidewalk in shacks made of plastic paper. All of them said that they face water, sanitation, food and shelter problem. Besides, rickshaw pullers have to pay police. Others like street vendors have to pay local mastans and police. Women living at home take care of the children and the family, livestock etc. They have to face different problems (health, food etc.) all by themselves. Sometime they do not get money from their male partners on time. They have to loan from local Mahajans on high interest. They also mentioned about getting proper work in cities.

Permanent migration People who have money they buy land outside the area and go permanently, but the percentages are very small. Some went to India and never came back. People go to cities, specially women for work in garment industry can only come home during festivals or long holidays. They only get one day off weekly. But some work overtime to make more money.

This year none migrated permanently as they wanted to stick together with their family and be secure and get help from each other. They said they are afraid of moving from their birth place and they do not have money to buy better land outside the area and they do not know other there.

None migrated permanently from this area. But from 20% household members go to cities to work in the garment industries for longer time.


16 days 20 days 15


Cattle were moved to higher ground such as school ground, embankment and higher land and remained there for 16 days until the homestead dries up. Very few livestock died People usually keep cattle as safety net. In 2016 they fattened cattle before flood and sold them during Muslim festival Eid-ul-Azha. But they could not get proper price as middlemen paid less knowing their inability to sell cattle by themselves due to flood and shelter. This is a common practice. As India restricted cattle movement from India to Bangladesh and there was huge demand for cattle during this festival, people fattened cows to sell.

They do not keep many cattle for longer time. Some of the household fatten cattle and sell them before monsoon. Some move their cattle to the protected area where there is higher ground, near school premises and on embankment and remained there for a month till they feel secured. About 1% livestock died and 30% were sold to recover from the after flood shock. It is not easy to maintain livestock in the chars during monsoon but during dry season there is plenty of grasses for the cattle. Households keep their cattle on the higher ground/boat. During flood they have to stall feed cattle but the feed they can provide was not enough and the livestock health was at stake. But they can sell livestock during at high prices

Almost all households keep livestock. When flood comes they move the livestock in safe places. 50% households sell their livestock during monsoon or before monsoon.


Chars Jatrapur federation Goruhara Gram Unayan Samity Porarchar Gram Unnayan Samity Status of embankment Within the adjacent chars there are 3 km

embankment of which 1km was breached in southern and northern sides in 2 places. There are 4 shelters (higher ground, killa) in this area

Western part of the embankment was breached in 2010 and Eastern part eroded in 2013. The breached parts have not been repaired.

There is no embankment

Initiatives to help people Early warning was given by different NGOs, and Government and Union Parisad. However, looking at the river current and rainfall in upper riparian areas char people realized that there will be mass disaster. They moved to higher ground in the main stable area. CBO has helped them to move.

Early warning by NGO and Government was provided. Embankment and schools were used as shelter. Government provided some relief although people complained it to be not enough.

In 2016 there was little erosion. The CBO members did not get receive much help as there was very little damage. They got help from the CBO

CBO role during flood/erosion One of the main activity of the CBO is disaster management. The CBO members received training on early management and formed volunteer group. They relocate people by using their boats and man power, find out space for the people and their livestock to move. They report to RDRS (a regional NGO) for relief and help. They also keep contact with the government departments for post flood/erosion rehabilitation. They provide early warning through own miking system. They also cooperate Union Parisad during relief distribution.

CBO members have helped people to move and build new houses or repair damaged ones. However, as the members are all poor and they have very limited funds they have not been able to do much. They prepared list of flood and erosion victims, appealed to NGO and government agencies for help for the victims. They also communicate with the Union Parisad.

The CBO has provided information on the flood and erosion to their members. In 2016 they provided Early warning and helping vulnerable households to move to the safer places by CBO boat before flood. They also provided water purification tablets and tablets for diahorrea.

Impact of embankment Normally embankment is used as shelter during flood and erosion and also used as road for communication. But this year embankment was breached in the adjacent char and flood water flooded the road cum embankment. As most of the char area is unprotected they have not been able to protect their standing crop (Aman rice, Mashkalai and peanuts) during 2016 flood. Sand deposition on the agricultural land also indicates that within next 2 years they will not be able to harvest better crops.

Embankment was built in 1996. Embankment have saved people before breach. There was not much flooding before and they had fish ponds, different crops and good irrigation system in the mainland before they moved to the new chars. . But now they have to move frequently due to erosion. Although embankment provides shelter to the flood and erosion victims, it also logs water inside the protected area when it rains a lot ad sometimes all standing cops go under water. Erosion victims build houses and live on the embankment. There is one shelter in the mainland area.

No embankment.

Disasters in last 10 years Flooding happens every year but was not so high. Flood affected in 2010, 2014 and in 2015 due to high rainfall. Erosion every year but severe in 2015 and

Flood causes misery every year but in more in 2013, 2014, 2015 and 2016. Erosion every year but severe in 2013. Drought: 2010, 2011 and 2015

In 2016 Dharla river current was stronger than in 10 years back, erosion and accretion are frequent. It takes 5-6 years to accrete land now in comparison to 30-40 years before.


Chars Jatrapur federation Goruhara Gram Unayan Samity Porarchar Gram Unnayan Samity 2016 Drought: 2011, 2012, 2013 and 2014 Excessive rainfall: 2010, 2015 Now a days the period of other natural calamities shifted a month ahead. Untimely rain, drought, fog, shorter winter and less winter rain are evident.

Excessive rainfall: 2014 and 2014 Other natural calamities decreased but thunderstorm and fog increased now. During 1998 flood homestead was not flooded but continuously there was standing water for 3 months on the agricultural land.

According to the respondents, flood in 2016 was little less than 1988 flood but river was violent in 2016. They suffer from heavy rain and strong wind in 2016. Flood waster stayed for longer time on the agricultural land. Flood usually came in August/September before but this year it started in July. Flood also affected the area every year but it was little higher in 2011, 2012, 2013 and in 2014 than other year. According to the respondents water level in the area was higher on the homestead in 2016 than in 2015. Drought affected the area in 2010 and in 2015/2016. Fog in high in 2014 caused severe damage to the winter crops.

Impact of flood Damaged houses, infrastructure, crops (Aman, Mash kalai, peanuts), seedbeds were under water, sand deposit on agricultural land

Damaged houses, infrastructure, crops (Paddy), seedbeds were under water. In exposed areas

Some people have to move several times. They have to dismantle their shacks and build it again. They lost all their agricultural land due to sand deposition. About 1.5 to 2ft sand deposited on the agricultural land. Some standing crops were buried under sand. , they have to sell their livestock.


Houses eroded, agricultural land eroded. As they have to stay in the open space they are exposed to rain and thunder shower. They have to move which caused financial loss. They have to sell some livestock and be separated from the original community.

Very few houses eroded but lost standing crops.

Coping with Flooding Loan from NGO, Mahajan, sell livestock, agricultural land, migrate for work for longer time


Loan from Federation and Mahajan, sold livestock, sold fixed assets, lease land out, migrate to work in other cities.



Loan from Federation and Mahajan, sold livestock, catch fish in the river, migrate for work in other cities.

Hydro-Social Deltas Migration and hazards 2016 A2-1

ANNEX 2 SURVEY FORMATS

NWO Hydro-Social Deltas Project

Flood and Erosion Responses 2016

Household Questionnaire

Name of CBO: _________________________________________________

Location: __________________________________________________

Type of site: 1- active floodplain char

2- between river and embankment

3- mainland (protected by embankment)

District: ……………………………. Upazila: ……………………………………………………

Union: ……………………………… Village: …………………………………………………….

A: Respondent Information

A1 Respondent name:………………………………………………………………………….......

A2 Father’s/Husband’s name ………………………………………………………………………

A3 Gender (circle): Male Female

A4 Age (circle): 18 – 30 31-40 41-50 51-60

A5 Education: Number of Years Completed: |__|__|

B: Household information

B1 How many members of this household are there (regularly present and eating together),

and how many if any live mostly away and send money home?

Male Female

Adult > 16

Child <16

Migrated and sending money

B4 What is your household’s main source of income? _________ |__|__|

B3. What is your household’s second source of income (if any) ? _________ |__|__|

Occupation codes:

1-cultivate own land 2-cultivate own and sharecrop land 3-sharecropper only 4-rent out land 5-fishing 6-fish trader 7-fish net maker 8-fish processing 9-fish culture 10-fish gear trader

11-agric labourer 12-non-agric labourer 13-rickshaw/van 14-boatman 15-handicraft 16-petty trade 17-business 18-mechanic/driver 19-other employee/ 20-teacher

21-government service 22-paid homestead work 23-housewife 24-livestock 25-poultry rearing 26-carpenter/mason/blacksmith 27-student 28-beggar 29-no activity other (specify)


B4. How much land (if any) does your household own and/or use now? (decimals)

Land Type In chars Between river and embankment

Behind embankment

Own homestead land

Own pond

Own cultivated Land

Other’s land share-cropped or leased in

Own land share-cropped or leased out

Own orchard/tree land

Other own land (fallow etc)

Other (not own) land used

*100 decimals = 1 acre

B5. Are you or anyone else in your household a member of:

any NGO group? Yes |__| No |__|

If yes, which: __________________________________________________

any CBO/samity? Yes |__| No |__|

If yes, which: __________________________________________________

C Flood and erosion 2016

C1 This year what losses if any did your household suffer due to erosion or flooding?

Type of loss Impact Flood Erosion Area (dec)

No house destroyed

Housing/ homestead loss

No house damaged

Crop/land loss Area

Cattle/buffalo no

Goat no

Livestock

Poultry no


Days lost/affected

No persons ill

Days affected

Health

Costs of treatment (Tk)



C2. Which of the following if any have you done to cope this year? Do you think you might need to do any of them in the future to cope if there is flooding or erosion? [tick if did/expect] Activity 2016 Future

1. Limit the portion size at meal times

2. Reduce number of meals

3. Borrow food

4. Eat cheaper or less preferred food

5. Reduce adult food consumption

6. Rely on casual labour for food

7. Migration for work

8. Migration for safety

9. Borrow money from NGOs

10.Borrow money from money lenders

11.Borrow money from friends and relatives

12. Sold harvest/produce in advance

13.Sold small animals

14. Sold large animals

15.Sold land

16.Pledging labour


D Migration in 2016 - Did your household move home in 2016 due to flood or erosion?

D1. Whole household moved and not returned to this year's starting place ("home") Yes-1, No-0 |__|

[complete for each location lived in as household, leave other rows blank - e.g. if moved twice this year then current location is location 2nd move to and no location 3rd move to. ] Location

(place name) Category Distance

from start

Why here Home-stead status

Main advantage there

Main disadvantage there

Risk Why left (moved)

When left

Starting

place this

year

0

Location

1st moved

to

Location

2nd move

to

Location

3rd move

to

category: 1-island char, 2-riverside (not protected), 3-on embankment, 4-village area behind embankment, 5-other more distant rural area, 6-town, 7-city distance: 0 = original location; 1 = <1 km; 2 = 1-3 km; 3=4-5 km; 4 = 6-10 km; 5 = >10 km why here: 1-birthplace, 2-own land here, 3-relatives here, 4-land available, 5-near home, 6-work available, 7-safe from flood, others- details... homestead: 1-own land, 2-public land, 3-rented land, 4-other person's land allowed to stay here without payment advantage: 1-productive land, 2-plenty of work, 3-safe location, 4-good public services (school or health etc), 5-good environment, 6-relatives and known people, 7-good communications, 8-land to live on, others - details.... disadvantage: 1-unproductive land, 2-little work, 3-risky location, 4-poor public services (school or health etc), 5-poor environment, 6-people not helpful, 7-poor communications, 8-no land to live on, 9-far from family/worry, 10-don't get paid properly, 11-don't get enough work, 12-theft/crime/tolls, 13-no/poor sanitation, 14-poor/no accommodation, 15-poor food, 16-high travel cost, 17-take loan on high interest, others - details.... risk to household wellbeing: 4-very high, 3-high, 2-moderate, 1-low, 0-none why left: 1-eroded, 2-flooded, 3-land owner/authority forced to go, 4-found better place to go, others-details.....


D2. Whole household moved and returned to this year's starting place ("home") Yes-1, No-0 |__|

[complete for each location lived in as household, leave other rows blank - e.g. if moved once this year then moved back to home, then home =

current location is 1st row, and location moved to as a household temporarily is location 1st moved to

Location (place name)

Category Distance from start

Why here

Home-stead status

Main advantage there

Main disadvantage there

Risk Why left (moved)

When left

When returned

Home

(starting

& ending

place)

0

Location

1st

moved to

Location

2nd

move to

category: 1-island char, 2-riverside (not protected), 3-on embankment, 4-village area behind embankment, 5-other more distant rural area, 6-town, 7-city distance: 0 = home/original location; 1 = <1 km; 2 = 1-3 km; 3=4-5 km; 4 = 6-10 km; 5 = >10 km why here: 1-birthplace, 2-own land here, 3-relatives here, 4-land available, 5-near home, 6-work available, 7-safe from flood, others- details... homestead: 1-own land, 2-public land, 3-rented land, 4-other person's land allowed to stay here without payment advantage: 1-productive land, 2-plenty of work, 3-safe location, 4-good public services (school or health etc), 5-good environment, 6-relatives and known people, 7-good communications, 8-land to live on, others - details.... disadvantage: 1-unproductive land, 2-little work, 3-risky location, 4-poor public services (school or health etc), 5-poor environment, 6-people not helpful, 7-poor communications, 8-no land to live on, 9-far from family/worry, 10-don't get paid properly, 11-don't get enough work, 12-theft/crime/tolls, 13-no/poor sanitation, 14-poor/no accommodation, 15-poor food, 16-high travel cost, 17-take loan on high interest, others - details.... risk to household wellbeing: 4-very high, 3-high, 2-moderate, 1-low, 0-none why left: 1-eroded, 2-flooded, 3-land owner/authority forced to go, 4-found better place to go, others-details.....


D3. One or more members of household moved separately (e.g. temporary, seasonal) from rest of household this year Yes-1, No-0 |__| [complete for each person and location involved. This is only for movement that did not involve the whole household.

No days away in month Person

sl M/F Age

(yrs) Yrs educ-ation

Location Category Region Work done here

Housing No return trips

N D J F M A M J J A S O Main disadvantage there

Risk

category: 1-island char, 2-riverside (not protected), 3-on embankment, 4-village area behind embankment, 5-other more distant rural area, 6-town, 7-city region: 1-same upazila, 2-same district, 3-neighbouring district, 4-further away (not Dhaka), 5-Dhaka why here: 1-birthplace, 2-own land here, 3-relatives here, 4-land available, 5-near home, 6-work available, 7-safe from flood, others- details... housing : 1-own building/land, 2-shelter on public land, 3-rented building, 4-rented room, 5-rented space in mess, 6-staying with relative without payment, 7-staying with friend without payment, 8-temporary shelter, 9-open space no return trips - number of times visited this place and came back home during year disadvantage: 1-unproductive land, 2-little work, 3-risky location, 4-poor public services (school or health etc), 5-poor environment, 6-people not helpful, 7-poor communications, 8-no land to live on, 9-far from family/worry, 10-don't get paid properly, 11-don't get enough work, 12-theft/crime/tolls, 13-no/poor sanitation, 14-poor/no accommodation, 15-poor food, 16-high travel cost, 17-take loan on high interest, others - details.... risk to wellbeing: 4-very high, 3-high, 2-moderate, 1-low, 0-none


E Hazards and movement in previous years

E1. In the last five years was your household impacted by any of the following?

No. of years in last 5 years Hazard Land/crop

loss Livestock Loss

Housing loss

Loss of work

Serious ill-health

Trend

1.Floods

2.River bank erosion

3. Waterlogging

4. Sand deposition

5. Infrastructure

development project

Record no of years with each loss in cells

Trend: 1-increasing, 2-about same, 3-decreasing, 4-can't tell/don't know

E2. If the household's land was acquired for infrastructure (e.g. embankment):

how much land (dec) |__|__|__|__|

which year |__|__|__|__|

E3. What types of assistance (if any) in response to floods and erosion have you received

this year (2016), and from what types of organisation? (tick just sources and types that this

households received, leave blank if received no such support)

Overall how useful was help from each source?

Type of help NGO Govern-ment

CBO Relatives and other individuals

Warning

Shelter

Housing afterwards

Food

Cash

Clothing

Loan

Work

Crop inputs

Protection from future events

Other (specify)

_________________

Rating of usefulness

Rating of usefulness: 4 - very good, 3 - good, 2 - average, 1 - poor, 0 - no use


E4. How do members of your household communicate with possible distant workplaces and

with members of the family when living away from home?

Method With sources of work With family at home With others

Face to face

Mobile phone calls

SMS

Letters

Other (specify)

Code: 2 - mainly use, 1 - use sometimes, 0 - not used

E5. How many times in your lifetime (respondent) have you had to move home due to?

Erosion/river moving |__|__|

Land reappearing/accreting |__|__|

Land acquisition for development project |__|__|

Evicted/forced out by land owner or authority |__|__|

F Opinions

F1 How do you rate the effectiveness of the following in helping cope with flood and erosion?

Rating

1. Knowledge and understanding of hazards

2. Early warning

3. Information on work opportunities and relief

4. Protection of home/land from flood and erosion

5. Flood free places for shelter and work

6. Micro-credit finance

7. Help to reconstruct house

8. Help to move home

9. Agricultural support packages

10.Cash help

11. Food aid

12. Health care

13. Law and order / security from theft

14. Other (specify) __________________________

Rating: 4 - very good, 3 - good, 2 - average, 1 - poor, 0 - no use


F2 How important are the following to your household's survival in 2016 [code]

Code Trustworthy relatives and neighbours in village

Savings

Keeping livestock

Embankment

Public building for shelter (e.g. school)

High land

Credit (NGO)

Credit (moneylender, shop..)

Help from CBO/samity

Mobile phone

Trustworthy persons in town or city

Access to natural resources (fish, grazing, ...)

Land to cultivate

TV/radio/newspapers

Code: 4 - Very important, 3- Important, 2 - Useful, 1- Little help, 0 - No use / don't have

F3. How has your household income and expenditure changed in the past 5 years?

Income source Income Expenditure 1. Crop farming

2. Fishing

3. Livestock

4. Wage labour (in this area)

5. Business /trade

6. Income from migration

Codes: 3 - Increased, 2 - Same, 1 - Decreased F4 How does your household compare now with 5 years ago?

Rating House structure

Days of work a year

Wage/income level

Savings

Land

Location living in

Ability to cope with risks and hazards

Codes: 5 - Much better, 4 - Better, 3 - Same, 2 - Worse, 1 - Much worse


NWO Hydro-Social Deltas Flood and erosion responses 2016

FGD checklist A Background A1 Name of CBO: _________________________________________________ A2 Location: __________________________________________________ A3 Type of site: active floodplain char between river and embankment mainland (protected by embankment) A4 No of villages covered:

Total _________

In char area _________

Behind embankments _________

Facing erosion _________

A5 Name of adjacent river ___________________________________ A6 What % of households living in the villages covered by this CBO are: A4a protected by embankments? _______ % A4b not protected by embankments? _______ % B Was the CBO covered in Gates project FGDs? If not, then use also additional module B checklist C This year C1 How much (%) of land in the area was eroded ? _______ % C2 How much (%) of land in the area was flooded ? _______ % Considering the area that was flooded, how long was it flooded for (% of flooded

area for each duration)? Up to 2 weeks _________ %

2-4 weeks _________ %

1-2 months _________ %

Over 2 months _________ %

C3 What % of homesteads were flooded? _______ % How many homesteads (if any) were lost to erosion? ________


C4 What types of movements of people, if any, have there been from this area? Type %

households Main reason Locations moved to What are the conditions

and concerns there How long they moved for on average

Evacuation – during hazard event

Other temporary movement

Seasonal movement

Permanent movement

Not moved

NA


C5 Overall how many households were involved in moving away due to

erosion _______

flood _______

C6 Distance of destination that people moved to and its risks Distance % of migrants Destination behind

embankments (%) Destination not protected by embankments (%)

Within CBO area

Adjacent villages (same Union

Elsewhere, same Upazila

Outside Upazila but inside District

Outside District (where?)

C7 What % of households (if any) moved livestock? Destination Average

duration moved for

Main reason Main outcomes (eg survived, sold, died)

Nearby higher place

Closest embankment

Further place

C8 Did any households move here from other places? If yes, where from? _____________________________________________

_________________________________________________________

why? _____________________________________________

_________________________________________________________

are they still here or moved back? ____________________________

_________________________________________________________

C9 What initiatives (if any) were there to help people cope (e.g. warnings, finding shelter,

moving, etc? _______________________________________________________________ _______________________________________________________________ Who took the initiative? _________________________________________ _________________________________________________________________


How did it work/what effect did it have? ______________________________ __________________________________________________________________ C10 Has there been any role of CBO in helping people cope (to stay here, or to move, or

other actions and needs? If yes, what - details _____________________________________ ___________________________________________________________________ How many households covered? _____________________________________ ___________________________________________________________________ Has this been helpful or not and how? ________________________________ ___________________________________________________________________ C11 What impact did embankments and sluices have for people of this area (if any) this

year? (within their home village or in the locations they moved to) ___________________________________________________________________ ___________________________________________________________________ ___________________________________________________________________ D In the last 10 years (before this year) D1 Were there significant floods or erosion and did people move home? Significant flood

or erosion (which)

What % people moved (temporary)

What % people moved (longer term)

Main reasons for moving

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006


D2 Destinations of migrants from here Distance % of

migrants Destination behind embankments (%)

Destination not protected by embankments (%)

Main risks here

Within CBO area

Adjacent villages (same Union

Elsewhere, same Upazila

Outside Upazila but inside District

Outside District

D3 Did any households move here from other places? If yes, where from? _____________________________________________

_________________________________________________________

why? _____________________________________________

_________________________________________________________

are they still here or moved back? ____________________________

_________________________________________________________

D4 What initiatives (if any) were there to help people cope (e.g. warnings, finding shelter,

moving, etc? _______________________________________________________________ _______________________________________________________________ Who took the initiative? _________________________________________ _________________________________________________________________ How did it work/what effect did it have? ______________________________ __________________________________________________________________ D5 Has there been any role of CBO in helping people cope (to stay here, or to move, or

other actions and needs? If yes, what - details _____________________________________ ___________________________________________________________________ How many households covered? _____________________________________ ___________________________________________________________________ Has this been helpful or not and how? ________________________________


___________________________________________________________________ D6 What impact did embankments and sluices have for people of this area (if any) this

year? (within their home village or in the locations they moved to) ___________________________________________________________________ ___________________________________________________________________ ___________________________________________________________________ E Overall E1 What advantages and disadvantages are there to each type of moving relevant here? Type of move Advantages Disadvantages

Temporary

Seasonal

Permanent


E2 Who benefits and who doesn't from moving, and why? E2a Benefits from moving (who and how) E2b Not benefit from moving (who and why not) E3 For people who moved to towns, how is their experience living there? (positive and

negative aspects), Do they face any hazards and risks? Positive Negative E4 Do you think the community here has been more or less able to cope with events this

year compared with previous years? how?, why?


E5 What were the main household coping strategies E5a this year? E5b in previous years? (note any differences)

Documents

World Programme - hydro-social-deltas.un-ihe.org...(Understanding flows of water and people to improve policies and strategies for disaster risk reduction ... 1.2 Flood Risks and 2016