Community-led AA for flash floods: Lessons learned from the last-mile community during 2022 Extreme Event in North-Eastern Bangladesh

Forecast-Based Early Action (FbA) is a promising disaster risk reduction technique that allows communities to take proactive steps with the help of accurate forecasting before a disaster strikes. The current global evidence indicates that timely FbA can save more lives and minimize the impact on communities in the emergency and recovery stages. However, the FbA funded by humanitarians or governments needs some specific forecast window (e.g., 7 to 9 days for riverine floods in Bangladesh) from impact identification to intervention deployment. But in the case of rapid on-set disasters (such as flash floods (FF)), such forecast windows might be difficult to identify as these disasters might happen within 5 to 6 hours. In such cases, our research focuses on how the last mile community takes anticipatory action (AA). As a case study site, we selected the north-eastern (NE) region of Bangladesh, which experienced extreme FF during June 2022.

The first goal of this study was to look into how flash floods change the impact dynamics of last-mile communities over time. The second goal was to investigate how forecasting can be improved in terms of effectiveness and inclusiveness. The third goal was to investigate community-led AA during normal and extreme FF events. To understand local experiences and observations related to climate and environmental cues, 12 Key Informant Interviews (KIIs) and 14 Focus Group Discussions (FGDs) were conducted during Nov-Dec 2023. The Key Informant Interviews (KII) were conducted with representatives from NGOs, CBOs, trade organizations, and government officials. FGDs were held with a variety of groups, including women, the elderly, the disabled, ethnicity, religion, and occupation.

Our research found that rather than official forecasting, communities rely on indigenous knowledge such as cloud patterns, wind flow, atmospheric changes in hilly areas, sudden water temperature drops, color changes, and so on. These indicators serve as early warning signs of impending flash floods, allowing residents to plan ahead of time. Based on these predictive indicators, they take proactive measures such as elevating house plinths and safeguarding essential assets related to their livelihoods around 2.5 months before the FF period. Because the global lead time for FF is short, any AA must rely on community action. Because the NE region of Bangladesh has a long history of FF, their solution would be beneficial for other parts of the world to learn about, especially as the world experiences more FF because of climate change.