The flash flood in 2017 in the northeast region, triggered by heavy rainfall, has breached parts of embankments and resulted in huge loss of asset and income of haor dwellers. The aim of the study was to investigate the impact of flash flood’17 on livelihood of haor dwellers and loss in agriculture- the major livelihood depending option. The study covered 70 villages of 31 unions under 6 districts (Maulvibazar, Sunamgonj, Sylhet, Habigonj, Netrokona and Kishorganj) with a total of 1845 households.  Both primary (quantitative and qualitative) and secondary data were used in this study. To detect the changes of monthly total rainfall for the period of 1987–2017, Mann–Kendall test was applied. The reasons behind this early flash flood in haor were identified as heavy rainfall within a very short period of time. According to the data from Indian Meteorological department (IMD), the rainfall data recorded in Meghalaya and Assam during the month of January, February and March which was this year respectively 29%, 40% and 73% higher than the normal rainfall (average of the rainfall over a 30-year period) of the corresponding months. Therefore, the recorded rainfall was 339.6 mm in March in Sylhet, which was 154% higher than the normal rainfall in March. Local communities also reported that insufficient protective measures including repair and maintenance of embankments, timely not completion of project works were some other reasons of huge loss.

As the results, agriculture was found as the main occupation of the Haor dwellers (about 52% household) in the normal period. But only about 13% of people were engaged in agriculture during the flood; the highest about 24% people were engaged in fishery at the same period of time. Interestingly, almost all people were occupied in any of the professions in normal period, but during the flood, a great number people (about 21%) stayed without job in the study areas. Respondents reported that they changed their usual food habits, roughly one-third of households consumed two instead of three meals per day, 53% adapted consuming to less food and 16% had low nutritious food due to the scarcity of food. Overall, one-third of people were found food insecure after the flash flood as per Household Food Insecurity Access Scale (HFIAS). Status of insecurity was highest in the Sunamgonj district where most of the area was affected. For continuing education, lack of transports/transportation facilities was found as the main barrier behind students’ irregular school. Overall, 92% students reported that they attended school 6 days per week before flood, which was dropped to about 63% during flood. The highest absentees were found in the study area of Moulivibazar because they were not accustomed to usual flood. Due to less accessibility to the market (e.g. disruption in transportation, no road connectivity, far away from the house, etc.), household members were not able to sell their products timely and with right price to maintain their livelihood in a way. To maintain livelihood after the flood, over two-thirds of households took loan from their relatives (42%) and NGOs (20%). Although no significant changes in access of tubewell water and improved sanitation in the study area, however, the prevalence of water-borne disease were found higher after the flood. The study also estimated total loss in agricultural sector including crop, poultry and dairy and day labored (BDT 29180 million) and health related expenditure (BDT 240.0 million) due to flash flood in haor areas. To reduce the vulnerability and improve livelihood of the victims, 22% and 76% of the victims received support from BRAC and Government emergency responses respectively within three weeks of flood. Almost 48% and 28% of supported households received BRAC and government aid within three weeks after the flood. This depicted that BRAC’s emergency response mechanism to the affected population was active and earlier in a humanitarian crisis like flood. Since the livelihoods with agricultural activities were being affected mostly in many years; the following policy recommendation can be taken to protect their assets and reduce loss from flood.