Physically Constrained Spatiotemporal Deep Learning Model for Fine-Scale, Long-Term Arctic Sea Ice Concentration Prediction

Accurately predicting Arctic sea ice concentration (SIC), especially during the melting season at subseasonal scales, is essential for advancing our knowledge of global climate change. Currently, statistical models for SIC prediction face three major challenges: limited spatial resolution and forecast timeliness, inadequate representation of sea ice dynamic processes, and difficulties in predicting SIC during the melting season. To address these challenges, this study introduces super-resolution SIC Transformer (SR-SICFormer), a deep learning-based, data-driven model designed for fine-scale, long-term Arctic SIC prediction. We also propose a novel physical constraints loss function, ConIce loss, which integrates thermodynamic and dynamic sea ice processes into the training procedure, aiming to improve the model’s predictive accuracy. Satellite remote sensing data are used for training and validating the model. Experimental results demonstrate that SR-SICFormer outperforms traditional statistical and numerical models in extended-range SIC prediction tasks, achieving a $5\times $ spatial super-resolution factor and a 15-day forecast period. The model achieves a root-mean-square error (RMSE) of 0.0504, a correlation coefficient (r) of 0.9420, a peak signal-to-noise ratio (PSNR) of 30.32 dB, and a structural similarity index measure (SSIM) of 0.9297. For the $2\times $ super-resolution and 60-day forecast during the melting season, SR-SICFormer maintains strong performance, keeping SIC residuals within [−0.3, 0.3]. In addition, the ConIce loss function effectively preserves both sea ice extent (SIE) and internal concentration distributions, ensuring that extended-range forecast results closely match ground truth in both SIE and concentration.