In a groundbreaking study, Chinese scientists have successfully predicted Antarctic sea ice levels for the December 2023 to February 2024 period using advanced deep learning techniques.
The research team, comprising experts from Sun Yat-sen University and the Southern Marine Science and Engineering Guangdong Laboratory in Zhuhai, employed a Convolutional Long Short-Term Memory (ConvLSTM) neural network to develop a seasonal-scale prediction model for Antarctic sea ice.
Their forecasts projected that Antarctic sea ice would remain near historical lows in February 2024, with the sea ice area (SIA) expected to be 1.441 million square kilometers and the sea ice extent (SIE) 2.105 million square kilometers. These figures were slightly above the record lows recorded in 2023.
Published in the journal Advances in Atmospheric Sciences in early February, the team’s predictions were validated using the latest satellite observations for February 2024. The observed SIA and SIE were 1.510 million square kilometers and 2.142 million square kilometers, respectively, demonstrating a remarkably close alignment with the predictions.
Researchers highlighted that the sea ice measurements from December to February fell within one standard deviation of the predicted values, underscoring the reliability of their forecasting system. \"Our successful prediction not only underscores the significance of strengthening Antarctic sea ice prediction research but also demonstrates the substantial application potential of deep learning methods in this critical area,\" said Yang Qinghua, a professor at Sun Yat-sen University.
The study marks a significant advancement in climate science, showcasing the potential of AI-driven models to enhance our understanding and forecasting of critical environmental parameters.
Reference(s):
Chinese scientists study precise Antarctic sea ice prediction
cgtn.com
