Our Privacy Statement & Cookie Policy

By continuing to browse our site you agree to our use of cookies, revised Privacy Policy and Terms of Use. You can change your cookie settings through your browser.

I agree

China boosts AI forecasting for extreme weather events

Sun Ye

China Meteorological Administration releases the
China Meteorological Administration releases the "Fengyuan" meteorological artificial intelligence model, Xiongan New Area, north China's Hebei Province, December 19, 2025. /CMA

China Meteorological Administration releases the "Fengyuan" meteorological artificial intelligence model, Xiongan New Area, north China's Hebei Province, December 19, 2025. /CMA

The China Meteorological Administration (CMA) released the "Fengyuan" meteorological artificial intelligence model Friday in Xiongan New Area, Hebei Province.

The model provides an open-source platform to enhance AI weather forecasting and advance meteorological research. CMA Deputy Director Cao Xiaozhong said "Fengyuan" marks a significant milestone, as the meteorological sector now has an independently developed, end-to-end model with proprietary intellectual property rights, achieving technological autonomy in key areas.

Wang Yaqiang, chief scientist of the development team, said the long-term goal is to create an observation-driven, multi-sphere forecasting model that integrates physical mechanisms while remaining open to attract industry experts and consolidate research efforts.

The model is expected to enhance extreme weather detection and serve sectors including low-altitude economy, energy security, transportation, and health meteorology. It will also serve the countries who have participated in the Belt and Road Initiative.

The CMA also announced upgrades to three existing AI forecasting models on Friday. "Fengqing," a global medium-range forecasting system, added 11 physical parameters including precipitation and solar radiation. "Fenglei," a nowcasting system, upgraded from radar echo prediction to direct quantitative precipitation forecasting with improved accuracy for extreme rainfall. "Fengshun," a subseasonal-to-seasonal prediction system, enhanced its spatial resolution and improved temperature and precipitation predictions.

These upgraded models will provide stronger support for disaster warning, energy management, agricultural planning and international climate services, according to the CMA.

Search Trends