Huawei Cloud And Shenzhen Meteorological Bureau Announce Regional AI Model For Weather Forecasting

On Tuesday, March 26, 2024, in Shenzhen, China, a significant advancement in weather forecasting was unveiled as Huawei Cloud and the Meteorological Bureau of Shenzhen Municipality announced the official deployment of their regional AI weather forecasting model. This innovative model, known as the "Zhiji" Regional Model, stands out for its ability to produce accurate five-day weather forecasts within a precise 3 km range. The announcement coincided with World Meteorological Day 2024 celebrations in Shenzhen.

The Zhiji Regional Model is based on Huawei Cloud's Pangu-Weather Model and has been pre-trained on high-quality regional datasets. It marks a significant step forward in meteorological science by providing forecasts for Shenzhen and its neighboring areas with an unprecedented spatial resolution of 3 km. This is a notable improvement over traditional global models, which typically operate with a 25 km spatial perimeter. The model covers various meteorological elements such as temperature, precipitation, and wind speed.

During its trial phase last month, the Zhiji Model demonstrated exceptional accuracy in predicting several cold-temperature periods. Plans are underway to further enhance the model's capabilities, particularly in providing precise precipitation forecasts. Lan Hongping, Deputy Director of the Meteorological Bureau of Shenzhen Municipality, emphasized the critical role of weather forecasting in disaster prevention and mitigation. He highlighted the bureau's ongoing efforts to improve weather services through AI innovations in collaboration with Huawei Cloud.

William Dong, President of Huawei Cloud Marketing Dept, remarked on the significance of this development. He noted that the launch of the Zhiji Regional Model signifies a new era where AI can facilitate more accurate and smaller-scale weather forecasts. Dong also referenced a paper on the Pangu-Weather Model published in the renowned scientific journal Nature in July 2023, showcasing how scientific discoveries can extend beyond the laboratory to offer practical benefits.

The World Meteorological Organization reports that weather, climate, and water-related hazards have led to nearly 12,000 disasters between 1970 and 2021, causing over two million deaths and $US 4.3 trillion in economic damages. Early warnings can drastically reduce casualties and prevent economic losses. AI's ability to process vast amounts of data and identify patterns enhances the precision and speed of weather forecasting.

With extreme weather events becoming more frequent worldwide, AI weather forecasting systems have proven their unique strengths in predicting various weather conditions. Huawei Cloud plans to continue innovating in AI weather forecasting methods to benefit more sectors, provide refined weather services, and improve disaster preparedness.

In July 2023, a significant paper detailing Huawei Cloud's Pangu-Weather Model was published in Nature. Following this, the model was made publicly available on the website of the European Center for Medium-Range Weather Forecasts (ECMWF) in August and was named a top 10 scientific achievement in China for 2023. In December 2023, Huawei Cloud announced a collaboration with the Thai Meteorological Department to develop a Pangu-Weather Model for Thailand.

As Southern China approaches its monsoon season, Huawei Cloud and the Meteorological Bureau of Shenzhen Municipality are set to further verify and comprehensively evaluate Zhiji during this period. Their continued efforts aim to enhance the model's performance and provide valuable insights for weather forecasters.