Full Breakdown
Hong Kong Scientists Develop AI Model for Advanced Weather Forecasting
1/28/2026, 9:32:16 PM
Breakthrough in Weather Prediction
A team of scientists from the Hong Kong University of Science and Technology has developed an innovative artificial intelligence (AI) model designed to enhance the prediction of thunderstorms and heavy rainstorms. This new system, known as the Deep Diffusion Model based on Satellite Data (DDMS), can forecast severe weather events up to four hours in advance, significantly improving upon the current prediction range of 20 minutes to two hours. The model utilizes satellite data and generative AI techniques, achieving over a 15% increase in forecasting accuracy.
Collaboration and Methodology
The DDMS was developed in collaboration with China's national meteorological agencies and is based on infrared brightness temperature data collected by China's Fengyun-4 satellite between 2018 and 2021. This satellite technology allows for earlier detection of cloud formations compared to traditional radar systems. The model incorporates meteorological expertise to effectively capture the evolution of convective cloud systems, which are critical for predicting rainfall.
Dr. Su Hui, chair professor of civil and environmental engineering at the university, emphasized the importance of this advancement, stating, “We hope to use AI and satellite data to improve prediction of extreme weather so we can be better prepared.” The model's ability to refresh forecasts every 15 minutes positions it as a vital tool for governments and emergency services in responding to increasingly frequent extreme weather events linked to climate change.
Context of Weather Extremes
The development of the DDMS comes in response to a notable increase in severe weather incidents in Hong Kong and southern China. In 2025, the city issued its highest rainstorm warning five times and the second highest 16 times, surpassing seasonal norms. This trend highlights the urgent need for improved forecasting systems to enhance regional disaster preparedness and response.
Official Statements & Responses
Dr. Dai Kuai, a postdoctoral fellow involved in the project, remarked on the significance of the model, describing it as a “major advancement in atmospheric monitoring and severe weather early warnings.” The collaboration between Hong Kong and mainland Chinese scholars underscores a collective effort to bolster disaster response capabilities through technology.
Criticism & Opposition
While the advancements in forecasting technology are largely welcomed, some experts caution against over-reliance on AI models. Concerns have been raised regarding the potential for inaccuracies in rapidly changing weather conditions, emphasizing the need for continued human oversight in meteorological predictions.
What's Next
The integration of the DDMS into official forecasting systems is underway, with both China's Meteorological Administration and Hong Kong's Observatory working to incorporate this model into their operations. As climate change continues to influence weather patterns, the need for such advanced predictive tools will likely grow, making this development a crucial step in enhancing public safety and preparedness.
