Google researchers leverage AI technology to forecast floods in India, potentially saving lives

With the recent boom in artificial intelligence (AI), concerns have arisen about this technology’s ability and potential to do good as well as bad. While fear is being expressed at the highest levels about the high potential of the risk involved in this technology, there is no stopping this particular Juggernaut! And, looking to leverage AI for the good of humankind is Google as is evidenced by its recent collaboration with Apollo Hospitals in India to support early disease detection. And now, researchers are tackling another issue - predicting floods. They hope to forecast floods by leveraging AI technologies, potentially saving lives.

Also Read: Microsoft onboards former Google DeepMind co-founder to head AI products including Copilot and Bing

Google researchers harness AI technology to forecast floods

In a research paper published in Nature, Google researchers shed light on how they are harnessing AI to work towards their goal of flood prediction. Machine Learning (ML) is being used to create hydrological simulation models that can be transferred to basins that are yet to be gauged. For this, researchers say they are using Global Flood Awareness System (GloFAS) version 4 which the European Centre operates for Medium-Range Weather Forecasts (ECMWF). Comparing GloFAS to the latest river prediction models developed by the researchers reveals that it not only provides accurate predictions earlier but also covers “larger and more impactful events.”

Yossi Matias, VP of Engineering & Research, and Grey Nearing, Research Scientist, Google Research said, “The flood forecasting initiative is part of our Adaptation and Resilience efforts and reflects Google's commitment to address climate change while helping global communities become more resilient. We believe that AI and ML will continue to play a critical role in helping advance science and research towards climate action.”

Also Read: Google Gemini AI images disaster - What really happened with the image generator?

The AI model used in the study predicts daily streamflow across a 7-day forecast horizon using long short-term memory (LSTM) networks. Researchers say this model was trained from a sample across 5680 streamflow gauges using “random k-fold cross-validation”. This means the Flood Hub system can deliver river predictions up to 7 days in advance, covering rivers across 80 countries.

In their multi-year journey, researchers have “advanced research over the years hand-in-hand with building a real-time operational flood forecasting system that provides alerts on Google Search, Maps, Android notifications and through the Flood Hub.”

One more thing! We are now on WhatsApp Channels! Follow us there so you never miss any updates from the world of technology. ‎To follow the HT Tech channel on WhatsApp, click here to join now!