Flash Floods can be difficult to predict, and when such a disaster strikes, information can be a lifeline. Google is now taking steps to address this with a new Gemini AI-powered flash flood prediction tool, Groundsource. The new platform uses Gemini to source data from old news reports. This is definitely a step forward in flash flood prediction, as it can help with better disaster preparedness.
Google says that Groundsource used Gemini brains to analyze decades of public reports and found over 2.6 million historical flood events spanning more than 150 countries. It then used Google Maps to pinpoint precise geographic boundaries for each event to build a dataset focused on flash floods. Using this dataset, the company trained a new model that makes “tangible progress” towards predicting flash floods in urban areas up to 24 hours in advance.
“By turning public information into actionable data, we aren’t just analyzing the past — we’re building a more resilient future for everyone towards our goal that no one is surprised by a natural disaster,” notes the tech giant.
Google says urban flash flood forecasts are available in Google’s Flood Hub, along with the existing riverine flood forecasts. It apparently covers 2 billion people in more than 150 countries.
Flash flood forecasts are available in Google’s Flood Hub
Google is also sharing its data with emergency response agencies in these locations. This same AI approach of the Groundsource has the potential to predict other natural disasters, like landslides or heat waves, says Google.
That said, there seems to be some limitations. The model can reportedly only identify risks across a 20-square-kilometer area. Also, Google’s model doesn’t integrate local radar data. This means it may not be as precise as the US National Weather Service’s flood alert system. This data typically enables real-time tracking of precipitation. However, the new platform is specifically designed to work in regions that don’t have access to advanced weather-sensing infrastructure.
