The Hong Kong Observatory and mainland China’s National Meteorological Centre recently adopted an advanced AI model to better predict when typhoons rapidly intensify, a previously unpredictable weather phenomenon that poses significant challenges for disaster preparedness.
- AI model predicts rapid typhoon intensification with higher accuracy
- Hong Kong forecasts 4-7 typhoons this season, some may become super typhoons
- Chinese and US meteorological agencies advance AI-driven weather forecasting
What happened
A sophisticated AI model was deployed at the Hong Kong Observatory and China’s National Meteorological Centre about three weeks ago, designed to tackle the difficulty in forecasting rapid typhoon intensification. Led by Professor Li Qinglan of the Shenzhen Institutes of Advanced Technology, the AI system successfully provided real-time updates on Typhoon Jangmi’s progression, which impacted flight schedules in Hong Kong.
This AI model applies multiple machine learning techniques to predict when a tropical cyclone’s wind speed will ramp up sharply within short periods, an aspect historically very tough to forecast accurately. Testing over recent years demonstrated its improved performance over existing methods, particularly compared to US forecasts from the National Hurricane Center.
Why it matters
Rapid intensification events are rare and difficult to predict, often leaving communities and emergency services with limited time to prepare. By improving the accuracy and timeliness of these forecasts, the new AI model significantly enhances disaster preparedness and response efforts, which are critical as Hong Kong faces a forecast of four to seven typhoons between June and October.
The likelihood of super typhoons development this season is heightened due to the El Nino phenomenon, making improved forecasting tools indispensable. This advancement not only benefits Hong Kong but also contributes to regional meteorological capabilities, as accurate typhoon forecasting can reduce economic disruption and save lives.
What to watch next
Ongoing evaluation will focus on integrating real-time data to refine the AI model’s forecasting accuracy and operational performance. The Hong Kong Observatory plans to continue enhancing AI applications across its forecasting platform, making some products publicly accessible to increase community awareness and preparedness.
The development of AI for typhoon and broader severe weather prediction remains a priority in China and internationally, with collaborations such as the US National Hurricane Center partnering with Google DeepMind. Future advancements may extend forecasting lead times and improve predictions for other extreme weather events like strong winds, heavy rainfall, and storm surges.