Google has introduced a new AI-powered methodology called Groundsource that is aimed at improving the prediction of flash floods in urban areas. The update expands the capabilities of Google Flood Hub, allowing the system to forecast urban flash floods up to 24 hours in advance. The announcement was highlighted by Sundar Pichai, who said the company trained a new flood forecasting model to address the long-standing lack of high-quality data on flash floods. The model is designed to improve disaster preparedness by providing earlier warnings for communities.
According to Google, Groundsource is built using Google Gemini to analyse decades of public reports and historical information. The AI system identified more than 2.6 million historical flood events across over 150 countries, creating a large dataset that can be used to train flood forecasting models.
To improve geographic accuracy, the system used Google Maps to determine the boundaries of each event. This helped build a structured dataset focused on flash floods, particularly in urban environments where such disasters can occur quickly and with limited warning.
Google said the dataset is being made available to researchers and partners as an open-source benchmark to support further work in disaster prediction and climate resilience. The new dataset enabled Google to train a model capable of predicting urban flash floods up to 24 hours before they occur. These forecasts are now available through Flood Hub alongside Google’s existing riverine flood forecasts.
See What’s Next in Tech With the Fast Forward Newsletter
Tweets From @varindiamag
Nothing to see here - yet
When they Tweet, their Tweets will show up here.
