In Bangladesh, the monsoon season puts up to a third of the country under water. For weavers like Lal Chand Miah, with handlooms typically located on the ground, receiving early warning information is an absolute necessity in order to prevent financial losses from damages brought by rising floodwaters.
A pilot project, led by Deltares in partnership with HKV Consultants and with technical support from RIMES, allowed communities in Sirajganj to use their mobile phones to receive voice flood warning messages from the Flood Forecasting and Warning Center (FFWC) of the Bangladesh Water Development Board (BWDB), as well as send water level observations to FFWC via SMS. The latter makes them active participants in the flood early warning system, and not just recipients of forecasts and warnings. This initiative built on the CARE-supported project that RIMES implemented from 2012-2014, which expanded the geographical coverage of the CFAN 10-day flood forecasting model, delivered forecast and warning information to pilot communities via SMS, and built capacity of disaster management officials and communities in using these information and in making and recording water level observations. Use of voice messages in the Bangla language for early warning was one of the key feedbacks received from these pilot communities.
Trained gauge readers recorded water level observations five times a day, and sent these once daily by SMS to FFWC. Observation data were assimilated into the forecast model, for generating water level forecasts in major river systems. Forecasts and observation data were then interpreted to generate location-specific water level forecasts, which were then disseminated as voice messages with respect to current water levels, to 300 participants as direct recipients. Voice messages were sent once a day during the flood season, and twice a day when current water level is above danger level and water level forecasts show a rising trend. Community volunteers were trained in interpreting forecasts and responding to warnings.
Evaluation of the pilot project revealed that forecast and warning messages reached about 30,000 people in Rajapur and Ghorjan Unions (about 45% of combined population) as direct and indirect recipients (through people who directly received the messages). Of these, 80% were found to have understood the messages. Recipients took appropriate actions in response to these messages, such as moving livelihood assets to higher ground, harvesting early, or delaying planting. This led to savings (in terms of avoidable damages) at an average of USD 472 per household. Savings were highest in households engaged in fishery at USD 768, followed by those engaged in livestock rearing at USD 678 and in crop production at USD 640. Demand for increased frequency in message delivery and at specific time of the day to enable response was also noted.
Left: Abu Sayeed, fisherman, Rajapur, Belkuchi. Abu saved his hatchery worth USD 2,500 due to early flood warning, by enclosing his pond with net to prevent fish from being washed away.
Right: Samad Miah, farmer, Ghorjan, Chowhali. Samad delayed onion planting in response to forecast and avoided potential production losses.
This successful pilot has led to a follow-up project, supported by Cordaid, aimed at increasing availability of and access to risk and warning information through an online interactive web portal, linked to mobile services, at BWDB District Flood Information Centres and Union Parishad Digital Centres, matched with capacity building and public awareness activities.