Google has announced the expansion of Flood Hub, its AI-enabled flood prediction platform, to 60 countries in Africa, the Asia-Pacific region, Europe, and South and Central America. Flood Hub allows governments, aid organizations, and individuals to access flood forecast information up to seven days in advance.
Flood Hub specializes in forecasting riverine floods, which occur when rivers or streams burst their banks and flood nearby regions. Its influence has spread to the Central American “dry corridor,” which includes Nicaragua, Honduras, and Guatemala.
It will combine two existing forecasting algorithms to precisely calculate the volume of water entering a river, predict impacted regions, and estimate flood depth. It also focuses exclusively on riverine floods, leaving people vulnerable to flash floods, urban floods, and coastal floods without the same amount of early warning.
Yossi Matias, Google’s VP of Engineering & Research and Crisis Response Lead, emphasizes the platform’s reliance on AI’s immense data processing capabilities, enabling the analysis of vast amounts of information within seconds. The models employed by Flood Hub show significant improvement as they train on additional data, mitigating gaps in historical flood records in certain regions.
The sources for this piece include an article in Axios.