Scientists developed AI Vakula: Predicts storms 4 hours in advance

“We hope to be able to use AI and satellite data to improve extreme weather forecasting so we can be better prepared,” said Su Hui, a professor in the university’s Department of Civil Engineering and Environment, who led the project.

The purpose of the system is to predict heavy rains, Su said at a news conference called to describe the paper published in the journal Proceedings of the National Academy of Sciences in December.

Developed in collaboration with Chinese meteorological authorities, the system updates its forecasts every 15 minutes and has increased accuracy by over 15 percent, the team said.

Their work is key for Hong Kong and China, which faced an above-average number of typhoons and rains in 2025, scientists said.

Last year, the city issued the highest rain warning five times, and the second highest warning 16 times, which has never happened before, said the Hong Kong meteorological authorities, which are working together with the Chinese to apply the new model to their forecasts.

DDMS was trained on temperature data collected between 2018 and 2021 by China’s Fengyun-4 satellite.

Satellites can detect cloud formation earlier than other forecasting systems such as radar, Su added.