Google uses AI and 5 million news reports to predict flash floods across 150 countries

Google is using old news reports and AI to predict flash floods

Flash floods are among the deadliest weather events in the world, killing more than 5,000 people each year. They're also among the most difficult to predict. But Google thinks it has cracked that problem in an unlikely way -- by reading the news. While humans have assembled a lot of weather data, flash floods are too short-lived and localized to be measured comprehensively, the way the temperature or even river flows are monitored over time. That data gap means that deep learning models, which are increasingly capable of forecasting the weather, aren't able to predict flash floods. To solve that problem, Google researchers used Gemini -- Google's large language model -- to sort through 5 million news articles from around the world, isolating reports of 2.6 million different floods, and turning those reports into a geo-tagged time series dubbed "Groundsource." It's the first time that the company has used language models for this kind of work, according to Gila Loike, a Google Research product manager. The research and data set was shared publicly Thursday morning. With Groundsource as a real-world baseline, the researchers trained a model built on a Long Short-Term Memory (LSTM) neural network to ingest weather global forecasts and generate the probability of flash floods in a given area. Google's flash flood forecasting model is now highlighting risks for urban areas in 150 countries on the company's Flood Hub platform, and sharing its data with emergency response agencies around the world. António José Beleza, an emergency response official at the Southern African Development Community who trialed the forecasting model with Google, said it helped his organization respond to floods more quickly. There are still limitations to the model. For one, it is fairly low resolution, identifying risk across 20-square-kilometer areas. And it is not as precise as the US National Weather Service's flood alert system, in part because Google's model doesn't incorporate local radar data, which enables real-time tracking of precipitation. Part of the point, though, is that the project was designed to work in places where local governments can't afford to invest in expensive weather-sensing infrastructure or don't have extensive records of meteorological data. "Because we're aggregating millions of reports, the Groundsource data set actually helps rebalance the map," Juliet Rothenberg, a program manager on Google's Resilience team, told reporters this week. "It enables us to extrapolate to other regions where there isn't as much information." Rothenberg said the team hopes that using LLMs to develop quantitative data sets from written, qualitative sources could be applied to efforts to building data sets about other ephemeral-but-important-to-forecast phenomena, like heat waves and mud slides. Marshall Moutenot, the CEO of Upstream Tech, a company that uses similar deep learning models to forecast river flows for customers like hydropower companies, said Google's contribution is part of a growing effort to assemble data for deep learning-based weather forecasting models. Moutenot co-founded dynamical.org, a group curating a collection of machine learning-ready weather data for researchers and startups. "Data scarcity is one of the most difficult challenges in geophysics," Moutenot said. "Simultaneously, there's too much Earth data, and then when you want to evaluate against truth, there's not enough. This was a really creative approach to get that data."

Google Is Using AI, Historic News Coverage to Predict Flash Floods

Google has released a new AI-curated dataset of news articles related to flooding that it claims can help predict when and where the next big floods will occur. With that dataset as a foundation, Google used an AI model trained on a Long Short-Term Memory (LSTM) neural network to analyze weather forecasts and make predictions based on historic news coverage, as TechCrunch reports. It can now estimate the probability of flash flooding in specific areas, and it's already helping emergency services respond more quickly. Despite mountains of data on rainfall, river flows, and global precipitation patterns, predicting flash floods has proved frustrating. They're so short-lived that gathering meaningful data is difficult, and historical data doesn't seem to provide enough correlation to suggest when and where they might occur. But news coverage is much broader and longer-lasting, and in cross-referencing flood reporting with weather data, Google thinks it's stumbled on the magic formula. The new Flood Hub service showcases areas with a high likelihood of flash flooding in the near future, with additional suggestions of probable flooding that are less certain. Google makes it clear that this tool is designed to augment, not replace, existing offerings. It can only identify risk areas that are 20 square kilometers, or around 12 square miles, which makes it far less accurate than the US National Weather Service's flood alert system, for example. But it also works in areas that organizations like this don't cover, worldwide. "Because we're aggregating millions of reports, the Groundsource data set actually helps rebalance the map," Juliet Rothenberg, a program manager on Google's Resilience team, said in a statement. "It enables us to extrapolate to other regions where there isn't as much information." This collaboration between hard data and news reporting is something Google wants to explore more as it finds new uses for its Gemini AI tool. Applying these points to other weather events, such as mudslides and heatwaves, may also be possible using this technique. They could also unlock important insights for the agriculture and construction sectors.

Google built a flash-flood prediction tool using Gemini and old news reports

Flash floods are , but Google might have a novel solution. The company , a prediction tool for flash floods that uses Gemini to source data from old news reports. This is the first time it has used a language model for this type of work. Google tasked Gemini with sorting through 5 million news articles from around the world and isolating flood reports. It transformed this data into a geo-tagged series of chronological events. Next, researchers trained a model to ingest current weather forecasts and leverage the Groundsource data to determine the likelihood of a flash flood in a given area. We don't have any concrete information as to how accurate Google's forecast model is, though that should come over time. One trial user did say it helped his organization respond quicker to localized weather events. For now, the company is highlighting risks for urban areas in 150 countries via its . Google is also sharing its data with emergency response agencies in these locations. There are some limitations here. The model can only identify risk across a 20-square-kilometer area. It's also not quite as precise as the US National Weather Service's flood alert system, because Google's model doesn't integrate local radar data. This data typically enables real-time tracking of precipitation. However, the platform's been designed to work in areas that don't typically have access to that kind of weather-sensing infrastructure. Juliet Rothenberg, a program manager on Google's Resilience team, hopes that this technology can eventually be used to predict other tricky phenomena. This includes stuff like heat waves and mudslides. "We're aggregating millions of reports," . "It enables us to extrapolate to other regions where there isn't as much information." This is Google's first use of a language model for weather forecasts, but not its first time it has relied on AI for this type of thing. The company's DeepMind has .

Groundsource: using AI to help communities better predict natural disasters

This content is generated by Google AI. Generative AI is experimental When disaster strikes, information is a lifeline. For years, as part of Google's Crisis Resilience efforts, we've provided early warnings about natural hazards to help communities stay safe. However, high-fidelity data for certain disasters like flash floods simply did not exist. This data gap has long prevented our ability to train AI models to predict flash floods before they happen -- until now. Today, we're introducing Groundsource, a new AI-powered methodology that transforms public information into a high-quality record of historical disaster data -- starting with flash floods in urban areas. Groundsource uses Gemini to analyze decades of public reports and identify over 2.6 million historical flood events spanning more than 150 countries. It then used Google Maps to determine precise geographic boundaries for each event to create a dataset focused on flash floods. Using this dataset we trained a new model that makes tangible progress towards predicting flash floods in urban areas up to 24 hours in advance. The urban flash floods forecasts are available in Google's Flood Hub -- along with our existing riverine flood forecasts which cover 2 billion people in more than 150 countries for the most significant riverine floods -- marking a significant expansion to our flood forecasting capabilities. For communities around the world, this means better preparedness before a disaster strikes. For our partners and scientists, Groundsource provides a massive, open-source benchmark to scale their impact -- particularly in urban regions that have lacked historical flash flooding data. Today, the Urban Flash Floods model and dataset join our Google Earth AI family of geospatial models and datasets. Importantly, the same AI-driven approach of Groundsource has the potential to be applied to other natural disasters, like landslides or heat waves, turning verified reports from around the world into datasets that enable improved global resilience. 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.