Season-Aware Hybrid Convolutional-Transformer for Antarctic Sea Ice Concentration Forecasting
2026-08-31 • Machine Learning
Machine Learning
AI summaryⓘ
The authors address the difficulty of predicting Antarctic sea ice concentration because it involves complex space patterns and long-term seasonal changes. They propose a new model that mixes convolutional layers (good at spotting local shapes) with transformers (good at understanding time relationships). They also add special techniques to teach the model about the monthly seasons to improve its accuracy. Their tests show this approach works better than other methods, especially when using these seasonal hints.
Antarctic sea ice concentrationConvolutional neural networksTransformerSelf-attentionSeasonal variabilitySpatio-temporal modellingPositional encodingTime series forecastingHybrid modelsAblation study
Authors
Danyang Li, John Taylor, Thang Bui, Quanling Deng
Abstract
Antarctic sea ice concentration (SIC) forecasting is an important yet challenging task due to the coexistence of complex spatial structure, long-range temporal dependencies, and strong seasonal variability. Conventional convolution-based models are effective at capturing local spatial patterns, but often have limited ability to model long-term temporal evolution. To address these challenges, we build on a hybrid Convolutional-Transformer forecasting framework for monthly Antarctic SIC forecasting. This framework combines convolutional encoding for spatial feature extraction with factorised self-attention for spatio-temporal dependency modelling. We further introduce two seasonal prior mechanisms: a month-aware positional encoding that injects calendar-month information into the token representation, and a seasonal temporal bias that encourages attention to periodically related historical states. Experimental results show that the proposed framework achieves better performance than convolutional and recurrent baselines across both classification and regression metrics. Ablation studies further indicate that the seasonal prior mechanisms provide consistent additional gains in both short- and long-horizon prediction. These results demonstrate the value of combining convolutional structures, attention mechanisms, and periodic prior information for Antarctic SIC forecasting.