ICTworks

The prediction and management of floods is a critical humanitarian issue in regions lacking hydrological data. Floods disproportionately affect developing countries due to the absence of dense streamflow gauge networks.

Subscribe Now for more digital development insights!

Global prediction of extreme floods in ungauged watersheds introduces an AI-based model that significantly improves the reliability of flood predictions, offering substantial benefits for humanitarians working in international development. This AI model achieves accurate predictions with up to a five-day lead time, surpassing the capabilities of current state-of-the-art systems that require extensive historical data for calibration.

Google Flood Hub for LMICs

One of the key advantages of the AI model in for developing countries is its global reach and accessibility. The model has been integrated into Google’s Flood Hub, an operational early warning system that produces real-time forecasts for over 80 countries. These technologies also enable Flood Hub to provide real-time river forecasts up to seven days in advance.

These forecasts are publicly available without any barriers such as cost or registration, ensuring broad accessibility. This open-access approach supports timely and actionable information dissemination, which is critical for effective flood management and disaster response by people, communities, governments and international organizations.

Flood Hub Humanitarian Impacts

For humanitarians working in international development, Flood Hub’s AI model’s ability to predict floods accurately and ahead of time offers significant benefits. Early warning systems are proven to reduce flood-related fatalities by up to 43% and economic losses by 35-50%.

By providing accurate and timely flood warnings, the AI model helps humanitarians plan and implement evacuation and other risk mitigation strategies more effectively, ultimately saving lives and reducing the impact of floods.

Flood Hub is also increasing the availability of hydrological data to further improve flood forecasts. The AI model’s design allows it to be used operationally without the need for local calibration, reducing the complexity and cost of deploying flood forecasting systems in resource-limited settings.

As the effects of climate change become more severe, floods often strike in unexpected places. Flood Hub research capabilities and technology will be able to forecast other types of flood-related events and disasters, including flash floods and urban floods.

Sending Cash Before Floods Happen

The International Rescue Committee and GiveDirectly are using $4.6 million from Google.org to send cash to 7,500 people in communities most exposed to an oncoming flood. The payments will go straight into their bank account within a day.

IRC and GiveDirectly found that giving families cash three or four days before a flood allowed them to purchase essential supplies like food and medicine, make reinforcements to their homes, relocate to safer areas, or reinvest in businesses or recover quickly after their crops were destroyed.