Supporting humanitarian anticipatory action in the Jamuna River basin, Bangladesh
What if you could forecast extreme flooding and the number of people who are potentially at risk from that flood event?
Using impact-based forecasts to take humanitarian anticipatory action and release pre-situated funding for the people and communities most impacted by flood events.
Early warning systems in Bangladesh
As more and more people across the globe find themselves vulnerable to natural hazards such as extreme weather and flood events, we’re working with the Start Network in Bangladesh to build analytical tools designed to support increasing resilience to flooding in a changing climate.
Using JBA’s unique Flood Foresight technology, we can not only predict the timing, location, extent and depth of flooding – we can also identify the number of people living in the impacted catchments and districts. Enabling local government and NGOs to trigger vital anticipatory action and emergency planning responses from in-country early warning systems to Disaster Risk Financing.
Since 2021, the Start Network, with its membership of over 90 aid agencies, has developed their ‘Start Ready’ programme. Start Ready pre-positions funding for crises that happen with regular and predictable patterns of recurrence, including floods. The Start Ready funding mechanism allows Start Network members to pool funds and risks, meaning they can stretch funding further than before – up to three times further than traditional humanitarian response funding. JBA’s flood risk and flood forecasting models and data provide the Start Ready programme with robust, industry leading analytics on which to trigger release of funding before flood events.
What did we do?
Working with the Start Network, we provided scientific modelling expertise to develop a real-time and forecast flood model for four districts on the Jamuna River basin - Kurigram, Gaibanda, Jamalpur and Sirajganj. The goal was to develop a system to estimate populations in each district that are forecast to be impacted by flooding at a range of timescales from 1-day to 10-days ahead.
How did we do this?
We used the JBA Global Flood Model to quantify flood risk to population through a probabilistic catastrophe risk model, resulting in outputs such as Annual Average Loss by district, Event Loss Tables and Exceedance Probability.
We then coupled the Copernicus Global Flood Awareness System (GloFAS) and our Global Flood Hazard Maps with our Flood Foresight technology to generate daily probabilistic forecasts of flood inundation extents and depths. These digital maps allowed us to estimate the population at risk of imminent floods throughout the wet season.
Dashboard technology
As part of this project we developed a bespoke online dashboard, which communicates the population forecast to be affected alongside the pre-defined triggers for the four districts. This fully automated impact forecasting system is providing the Start Network and its members with daily forecasts of flood conditions on which a series of anticipatory actions are being devised to reduce overall humanitarian impact.
Risk modelling and forecasting
Flood Foresight and the Global Flood Model both provide flexible analytical frameworks for integrating the best available data, including from local sources. Although this initiative was originally established based on global data and models, we continue to work with Start Network to integrate more locally-sourced data. This in turn improves the openness and acceptance of these models by in-country beneficiaries.
Our risk modelling and forecasting specialists are on hand to support the definition and operation of risk analytics in support of anticipatory action schemes globally. This is one of a number of risk and forecast modelling solutions we have deployed to support humanitarian anticipatory action, financial risk transfer and parametric insurance.
Reducing the overall impacts of flood and weather events is a key element of our ambition to lead in society’s pursuit of resilience to climate change.
