Google AI to boost PAGASA’s storm forecasting

Google’s artificial intelligence technology will now play a greater role in forecasting extreme weather in the Philippines under a strengthened collaboration between the Philippine Atmospheric, Geophysical and Astronomical Services Administration (PAGASA) and Japan-based Weathernews Inc. (WNI).

The partnership integrates Google’s flood forecasting and cyclone prediction models into PAGASA and WNI’s weather monitoring systems to enhance lead time, warning accuracy, and the precision of rainfall and storm impact projections across the country.

The Philippines, located along the Pacific typhoon belt, frequently experiences destructive storms that threaten lives and property. Authorities emphasized that improving weather forecasts is crucial to protecting vulnerable communities.

Under the collaboration, PAGASA and WNI forecasters now have access to what Google describes as “one of the world’s highest performing, innovative models,” allowing them to generate more complete assessments of a storm’s trajectory, intensity, and potential damage.

“We are pleased to see this partnership between WNI and PAGASA progress and hope our models can continue to make a positive contribution to forecasting in the Philippines,” said Raia Hadsell, VP Research, Google DeepMind.

Yossi Matias, vice president and GM of Google Research, said the initiative aligns with Google’s broader commitment to climate resilience. “It’s encouraging to see Google's AI models for flood forecasting and cyclone prediction being used in the Philippines. We believe that AI models, along with strong partnerships with governments and expert agencies like WNI, are key to building towards climate resilience, improving disaster preparedness, and making communities safer,” he said.

Google Philippines country manager Prep Palacios underscored the potential impact of the technology on public safety. “Improving the ability to forecast storms is a critical step in keeping Filipino communities safe. By bringing our AI models to PAGASA and WNI, we are giving forecasters a more complete picture of a storm's potential impact. We hope that this initiative will help improve disaster preparedness and build a safer Philippines, especially during typhoon season,” Palacios said.