China has begun trial operations of a new artificial intelligence system designed to improve sand and dust storm forecasting in some of the country's most hazard-prone regions.

The model, developed by scientists at the Chinese Academy of Meteorological Sciences' Lanzhou Institute of Arid Meteorology, began testing in Gansu province in late November, according to China Central Television.

Researchers raised the system's spatial resolution from 50 to 5 kilometers through a downscaling technique. The upgrade addresses a limitation in earlier 50-km AI products, which were built for large-scale regional dust tracking but lacked the precision required for localized forecasting and public-oriented services in Northwest China.

Speed is the other headline feature. Traditional physics-based models require supercomputers and long run times, which limit how often forecasts can be produced. The AI model runs on hardware using a graphics processing unit, or GPU, and generates global five-day predictions in under a minute. It updates eight times a day, giving meteorologists far more opportunities to track developing events.

"This is a major change, not only in speed but in accessibility. We can now run dust forecasts on a standard desktop computer," said Che Huizheng, a researcher at the academy.

The improvements were highlighted during a dust event in late November. Floating sand and dust formed in the southern Xinjiang basin, and blowing sand was reported in eastern Xinjiang Uygur autonomous region before the plume spread across parts of the Ningxia Hui autonomous region and provinces of Qinghai, Gansu, and other regions in North and Northwest China. The AI model detected warning signals two to three days ahead and issued a near real-time update early on Nov 22 that aligned closely with on-the-ground observations.

Yue Ping, a researcher at the institute, noted that spring remains the most active period for sandstorm outbreaks because exposed soil and frequent cold-air activity form a potent mix. Summer and autumn bring weaker events driven mainly by longrange transport, which are still difficult to forecast.

The system is intended to assist with both situations by providing higher-resolution optical and mass concentration data for multiple aerosol types, along with dozens of environmental indicators refreshed throughout the day.

Meteorologists say the project could offer a template for international cooperation in regions where dust storms regularly cross borders. Lower hardware demands may help agencies in Central Asia, North Africa, and the Middle East adopt similar tools for early-warning work.

Contact the writers at huyumeng@chinadaily.com.cn