Chinese scientist unveils AI model for weather forecasting at COP28

During the COP28 in Dubai, Chinese scientists from the Shanghai Academy of Artificial Intelligence for Science (SAIS), Fudan University, and China’s National Climate Center introduced a groundbreaking sub-seasonal forecasting model, “FuXi-Subseasonal,” utilizing artificial intelligence (AI). The model extends the forecasting range to 42 days, addressing a critical need in predicting sub-seasonal anomalies, which contribute significantly to high-impact climate events. Combining transformer architecture with guided random samples in a latent space, the model outperforms the European Centre for Medium-Range Weather Forecasts (ECMWF) in accuracy. Overcoming challenges related to inaccurate initial conditions, the model enhances predictions for the Madden-Julian Oscillation (MJO) and major weather processes, offering timely and precise assessments of climate-related risks. The researchers anticipate that FuXi-Subseasonal, with its cutting-edge AI capabilities, will contribute to addressing a range of climate challenges, including renewable energy development, electric power systems, agricultural planning, and sustainable socio-economic transformation.

Source – CGTN