The Observatory Organised a World Meteorological Organization Workshop on Recent Advances of Severe Weather Forecast Techniques using Machine Learning

19 December 2023

The Hong Kong Observatory (HKO) organised a workshop on "Recent Advances of Severe Weather Forecast Techniques using Machine Learning" under the Voluntary Cooperation Programme (VCP) of the World Meteorological Organization (WMO) from 4 to 8 December 2023.

This workshop was the first VCP workshop for HKO to resume in-person participation after the COVID-19 epidemic. It was also the first WMO training activity specifically on using artificial intelligence (AI) and machine learning (ML) in severe weather forecast. Invited speakers including Mr Jeffrey ADIE of NVIDIA, Dr Xuan YANG of the China Meteorological Administration (CMA), Prof Dit-yan YEUNG and Mr Chiu-wai YAN of the Hong Kong University of Science and Technology (HKUST), and Dr Ji-Hoon HA of the Korea Meteorological Administration (KMA), as well as Mr WONG Wai-kin, Mr WONG Ka-hing, Dr KWOK Yu-ting and Dr TAM Hiu-ching of the HKO, delivered lectures and shared their experience with the participants. In addition to the presentations, hands-on practical sessions on applying ML techniques in rainfall nowcasting and AI weather model for tropical cyclone track prediction were conducted.

The workshop was attended by 18 participants from the National Meteorological and Hydrological Services (NMHSs) in 18 countries, namely Bangladesh, Benin, India, Kuwait, Lao PDR, Lesotho, Madagascar, Maldives, Mongolia, Nepal, Nigeria, Oman, Pakistan, Singapore, Tanzania, Thailand, Uganda and Viet Nam. The WMO also appointed Mr Mustafa ADIGUZEL to join this workshop as an observer. He was delighted that this workshop was organised very timely by HKO with the theme well matching the needs of technique development for the WMO's high-priority initiative on early warnings. The workshop was also highly appreciated by the participants, and they were excited to see the potential benefits of AI/ML in weather forecast operations. Mr Nuthakit SINGHAPHET from Thailand commented that although the topic of machine learning was not familiar and challenging for some participants, new applications was covered in comprehensive details. Mr Usman ALI from Pakistan commented that the workshop was a great opportunity for him to learn about practical examples of AI applications in severe weather forecasting. Mr Kiwuuwa Joseph BALIKUDDEMBE from Uganda enjoyed the workshop and looked forward to more exchanges between domain experts in computer science and meteorology in the future. As concurred from the discussion sessions at the workshop, open-source AI weather models will make skilful data-driven weather forecasting increasingly common.

Public link

Please use the above public link if you want to share this noodl on another website.