PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
February 20, 2026Journal of Geophysical Research Machine Learning and Computation0 citationsOpen Access

IceCT: A MoE‐Based Model for Seasonal Arctic Sea Ice Concentration and Thickness Prediction

View Full Paper
ZZZ. ZhangYBYi BaoNCNing Chen

Key Points

  • To create a deep learning model that accurately predicts Arctic sea ice concentration and thickness.
  • Developed using a Mixture-of-Experts framework
  • Implemented Multi-Head Self-Attention for spatial pattern recognition
  • Utilized a Spatial Gate for expert contributions
  • Generated monthly forecasts at a 12.5 km resolution for up to 6 months ahead
  • Evaluated against deep learning baselines for performance comparison
  • Achieved a Mean Absolute Error of 4.56% for sea ice concentration
  • Recorded a Mean Absolute Error of 0.140 m for sea ice thickness
  • Demonstrated enhanced predictive skill during the summer melting period
  • Successfully mitigated the Arctic spring predictability barrier by relying on thickness information

Abstract

Abstract The rapid decline of Arctic sea ice requires accurate prediction of its concentration (SIC) and thickness (SIT). We introduce IceCT, a deep learning model using a Mixture‐of‐Experts (MoE) framework to generate monthly SIC and SIT forecasts at a 12.5 km resolution up to 6 months ahead. Its architecture features two specialized experts—one for SIC, one for SIT—enhanced with Multi‐Head Self‐Attention (MHSA) to capture spatial patterns. A novel Spatial Gate adaptively weights the experts' contributions across the grid. Evaluations show IceCT outperforms other deep learning baselines, achieving an overall Mean Absolute Error (MAE) of 4.56% for SIC and 0.140 m for SIT, and achieves enhanced predictive skill in forecasting sea ice during the crucial summer melting period. Additionally, the model mitigates the Arctic spring predictability barrier by learning to rely more on SIT information for forecasts initialized in spring.

Ask AI
Helpful
Bookmark
Share
View Full Paper

Cite This Study

Zhang et al. (2026) studied this question.

synapsesocial.com/papers/6997fa03ad1d9b11b3452ed0https://doi.org/10.1029/2025jh001010
Ask AI
Helpful
Bookmark
Share
View Full Paper