Google Expands AI Flood Forecasting Globally

"Valencia: 220 and rising. Hurricane Helene: 230. Hurricane Milton: 250. Storm Boris: 26. Rio Grande do Sul in Brazil: 176. Over 1000 victims across West Africa. I could go on. These are the people that have died in floods in the last three months with unprecedented rainfall."

These words, full of urgency and concern, come from Paul Simpson, a partner at ERM, who is vocal about the need for immediate actions against the deadly impacts of climate change worldwide.

One major player stepping up to meet this challenge is Google.

The tech titan recently unveiled a new strategy to tackle the growing perils of climate change: an ambitious expansion of its AI-enabled flood forecasting capabilities.

The initiative aims to extend its services to over 100 countries, potentially aiding approximately 700 million people. This is a dramatic increase from its prior reach, which covered around 460 million individuals.

Google's Chief Sustainability Officer, Kate Brandt, points out the escalating global concern about flooding. "Floods and storms were amongst the top natural disasters searched for in 2023, showcasing that the world is searching for this kind of information like never before," she states. With 2023 being a particularly harsh year for flooding, and 2024 already shaping up to be challenging, the Google team is poised to reduce the effects of these calamities. But what specifics does this new initiative involve?

Transformative Technology in Flood Prediction

Leading the discussion on this pioneering technology is Yossi Matias, Vice President & Head of Google Research. "Our mission is to use AI to make flood forecasting information globally accessible," he declares. "This expansion will bring critical flood information to millions of people across the globe, as well as more data to researchers and partners."

The improved model now boasts more accurate predictions, which hold for up to seven days — a notable enhancement over the previous model's five-day precision. This development has been made possible through cutting-edge research, which included the use of more labelled data and innovations in model architecture.

By tripling the number of locations used during the training phase and incorporating a new weather forecasting model, Google has substantially upped the robustness of its flood predictions. Such enhancements are timely, especially when recent disasters, like the catastrophic floods in Spain, were partly attributed to inadequate warning systems.

Broader Access for Research and Collaboration

Google is also planning to democratise access to its flood forecasting data by launching an API and the Google Runoff Reanalysis & Reforecast (GRRR) dataset in the near future. This will enable partners and researchers to tap into valuable hydrological forecasts and expected flood statuses, particularly in regions where local data is hard to come by.

Moreover, the introduction of "virtual gauges" promises to further broaden the availability of flood forecasting data in over 150 countries, providing vital insights even in locales without physical measurement infrastructure.

Learning from History to Safeguard the Future

In a move to aid historical analysis and long-term planning, Google has decided to make public the historical data outputs from its global flood forecasting model, which dates back to 1981. This treasure trove of data could help researchers grasp long-term flooding trends and craft strategies to mitigate future flood impacts.

Yossi Matias shares his vision for these applications: "We believe that access to this data will help researchers around the world understand how flooding is affecting communities, and potentially reduce the impact of devastating floods."

Indeed, the practical benefits of these advancements are already materialising. For instance, in May 2024, amid severe flooding in Rio Grande do Sul, Brazil, Google collaborated with the Geological Service of Brazil to unveil over 200 new monitored locations on its Flood Hub platform. This crucial information enabled local authorities and residents to develop effective emergency response strategies.

Early Warnings, Fewer Disasters

As climate change continues to amplify the frequency and intensity of extreme weather events, the role of advanced AI technologies in disaster preparedness becomes increasingly vital.

Google's enhanced flood forecasting capabilities dovetail with global efforts like the UN's Early Warnings for All programme, aiming to provide timely, actionable safety information for crisis response globally.

For Yossi, these developments are a source of hope. He expresses his aspirations: "When you think about other natural disasters... ideally, I’d like to get us to the place that nobody is surprised by any crisis or any sort of extreme weather event."

**************

Make sure you check out the latest edition of InsurTech Digital and also sign up to our global conference series for our sister site – FinTech LIVE 2024.

**************

InsurTech Digital is a BizClik brand.