A groundbreaking AI weather forecaster, Zhiji, has been unveiled by the team behind China’s Pangu-Weather prediction model. This innovative model has evolved from providing weather predictions with a precision of tens of kilometers to just a few kilometers. Zhiji focuses on regional weather and is capable of providing a five-day forecast with a precision sharpened from 25km to 3km. It was released less than a month after Pangu-Weather, developed by Huawei Technologies, was named China’s best scientific innovation of 2023.
Pangu-Weather first gained attention in July 2023 when a paper detailing the AI model was published in the journal Nature. It was launched on the European Centre for Medium-Range Weather Forecasts website a month later. One of its major milestones was completing a seven-day weather forecast in just 10 seconds, more than 10,000 times faster than traditional methods. In February 2024, Pangu-Weather was ranked first among China’s top 10 scientific advances in 2023 by the National Natural Science Foundation of China (NSFC). Not only does the model improve the ECMWF weather forecasting system, but it also provides 7-day predictions in 10 seconds.
The Pangu model delivers more accurate forecasts for crucial weather elements such as temperature, pressure, humidity, and wind speed compared to numerical simulations. Additionally, its error margin for predicting the paths of tropical cyclones is 25% lower than the ECMWF. By leveraging AI for weather prediction, researchers can bypass the complexities of traditional methods and provide more accurate forecasts.
Building upon the success of Pangu, the team has developed Zhiji in collaboration with the Shenzhen Meteorological Bureau. Zhiji has been trained with high-resolution data from southern China and can provide a five-day forecast with a precision of 3km for Shenzhen and its surrounding areas. While the Central Meteorological Bureau already provides hourly forecasts with street-level precision for the next 24 hours, Zhiji fills a gap in longer-term precision forecasting.
Currently, AI and manual predictions each have their strengths and weaknesses. AI excels in predicting the paths of typhoons, while numerical simulations are more accurate in determining wind strength values. A Huawei spokesperson stated, “Scientists can now integrate results from numerical simulations with forecasts provided by Zhiji to make the most advantageous judgments.” This integration of AI and numerical simulations could be a trend in the future of weather forecasting.
The true test for Zhiji 1.0 will come during this year’s flood season. Researchers expect the model to be further optimized with algorithm improvements based on the results. Ongoing work on the technology aims to enhance its rainfall forecasting capabilities and improve the resolution of heavy rainfall forecasts to 1km. In the future, precise meteorological models like Zhiji could provide early warnings for urban drainage systems in typhoon conditions by predicting street-level precipitation. Furthermore, if regional data from other areas are available for training, scientists could potentially develop local models tailored to those regions, serving more cities.
Overall, the unveiling of Zhiji marks another significant milestone in the transformation of weather forecasting through AI. With its increased precision and faster predictions, Zhiji has the potential to revolutionize how we prepare for and respond to extreme weather events. As researchers continue to enhance the capabilities of this technology, we can expect even more accurate and detailed weather forecasts in the future, helping us make better-informed decisions and protect lives and property.
Use the share button below if you liked it.