Attachments
Executive Summary
Regional Integrated Multi-Hazard Early Warning System (RIMES) is currently implementing “Supporting Flood Forecast based Action and Learning in Bangladesh (SUFAL)” project as a technical partner in collaboration with CARE Bangladesh, Islamic Relief Bangladesh and Concern Worldwide and funded by European Civil Protection and Humanitarian Aid (ECHO). This project focuses on reducing the vulnerability of flood-prone populations in the Brahmaputra-Jamuna basin by strengthening impact-based forecasting and early warning to trigger early actions and funding prior to flood events.
The project aims to emphasize and promote anticipatory or ex-ante actions that may minimize severe loss from hazards which maybe otherwise unavoidable. Despite Government of Bangladesh (GoB) having various policies and some guidelines in “Standing Orders on Disaster-2019” (SOD) depicting responsibility of different government bodies for anticipatory actions, post disaster response is currently more visible and in practice. However, changing from ex-post method to ex-ante activities can save and prevent massive loss of lives and livelihoods.
The objective of this assessment was to assess the reach of the forecast, advisories as well as their usability at community level, the impact of forecast based action during the monsoon of 2020 in SUFAL project areas and identify the gaps and areas of improvement for the implementation of forecast based action. SUFAL project piloted Forecast based Action (FbA) in Kurigram, Gaibandha and Jamalpur, three north-western districts of Bangladesh that get flooded almost every year. The study was conducted through onsite individual questionnaire survey, FGD and phone survey. A sample size of 669 participants was chosen for the assessment. Among which voice message recipients were 472 where 168 respondents were surveyed on site and 304 respondents were reached over phone. These included 197 participants in non-recipient group and 14 batches of focus group discussions also took place.
The flood of 2020 was comparatively of a longer duration than before and it reached the highest peak during the month of July. During the assessment, the severe impact of the flood was quite evident. The highest damage was incurred by the residents of Jamalpur districts 50,000 BDT per HH whereas in Kurigram and Gaibandha, most participants reported incurred damage ranging from 10,000 to 30,000 BDT per HH. Households incurred partial damages in household assets
(87%), livestock (84%), tube-well (81%) and sanitation (70%). Many of the households reported complete damages in agricultural crops (46%). In spite of following the advisories, some farmers could not harvest their jute because it was not ready for harvesting that time. In every district, more than 80% respondents answered they tried to relocate household assets to safer places and also tried to move to higher ground with their livestock. In Kurigram and Jamalpur, 46% and 60% respondents respectively who had crops on field that particular time experienced complete damage in crop. Except these deviations, people who responded immediately to early warning messages saved their crop. Most of them refrained from preparing seedbad or planting seedling after getting voice messages.
During monsoon 2020, voice message disseminated to around 8800 listed recipients including national and local government representatives, community representatives, UDMC, UzDMC, DDMC members, volunteers etc. The participants were regularly informed about flood situations and 98% of respondents reported that they received the flood Early Warning (EW), advisories or awareness messages. The perceived level of accuracy of voice messages reported by the community more than 93% in Gaibandha, 92% in Jamalpur and Kurigram has 67%. Again, not only perceived level of accuracy was high but also 90% of the recipients of both Gaibandha and Jamalpur said they understood the content of messages completely. Some people shared that people who did not get forecast information directly within the community were eager to know early warning from their neighbors who are recipients during FGDs. In Gaibandha, Jamalpur and Kurigram, 99%, 98% and 89% recipients respectively shared the voice message with neighbors. Respondents in Gaibandha (88%), Jamalpur (93%) and Kurigram (92%) responded that they took early actions after receiving flood Early Warning. Though the perceived level of accuracy of voice message was lowest (67%) in Kurigram, 92% people reported they took early actions after getting early warnings and it was possible due to the support from NGO, volunteers and local administration. In case of action taken, 45% participants took both individual and collective actions while 43% took individual actions. About 9% could not take any early action as some of them ignored the messages and some could not understand it while on the other hand, the aged people who didn’t have any supportive family members, also faced challenges in taking early action.
Respondents in Kurigram (63%), Gaibandha(39%) and Jamalpur (47%) responded that they got 1-3 days lead time for taking actions before the flood arrived while respondents in Kurigram(12%), Jamalpur (28%) and Gaibandha (15%) got more than 5days lead time to take action. To correlating the factors related to lead time, it has found that respondents who got 3-5 days or more than 5-days lead time, around 63% to 80% of them took immediate actions after getting the warning. The more people rely on advice from UDMC or wait for initiating actions of their relatives/ friends, the less lead time they could avail is found during the assessment. The percentage of these two factors is higher in the case of 1-3 days and less than 1 day lead time. This indicates that the more rapid the respondents started taking action, the more lead time they managed for taking early action.
Respondents reported that about 36% people had lack of capacity and resources for not taking early actions, 11% needed more lead time and did not availed lead time for taking actions, about 18% people had difficulties in the mobile network and 25% had no incessant electricity supplies. Despite some difficulties in taking actions, respondents took early actions according to their capacity and resources available to minimize loss and damage. Respondents of Gaibandha were able to save maximum in livestock saving of around 48,360 BDT per HH and about 32,180 BDT per HH which is highest in household assets as well as in agriculture about 22,358 BDT. On the other hand, Kurigram had the highest savings in the fisheries sector on an average 43,333 BDT per household. Overall, only the agriculture sector had the lowest savings in Jamalpur and Kurigram. There was not much crop to harvest and most of them refrained from planting Aman, could save lesser amounts in the agriculture sector and approximately 13,318 BDT per household. However, average savings per HH due to taking early action average cost on household assets were saved around 19,161 BDT per household, 36,552 BDT in livestock and 23,451 BDT in the fisheries sector. Though 2020’s flood peaked multiple times at an interval of 8 to 9 days and prolonged till September, in absence of the early actions, the damage stats would have been more.
In the pilot areas, about 10% respondents couldn’t bear the cost for taking actions, around 50% were able to bear the expenses and around 38% were partially able to bear the cost needed additional help. About 52% were completely able to manage the cost of early action despite
having constraints. Very few people got support from microcredit and FbF projects (SUFAL) during the flood while the Government support was also inadequate during the prolonged flood of 2020. SUFAL project used a Scenario based trigger approach which allows taking step by step actions according to expected severity of flooding. The impact of the voice message system was visible in all of the study areas. Even though they received the information, in many cases the participants were unable to act accordingly for multiple reasons. Summarizing the feedback from the community, need for longer lead time, addition of weather forecast and capacity building for forecast interpretation are the major requirements. Capacity building activities should also include decision making and systematic approach for early action adopting indigenous knowledge and practice could create better scope for early actions.