mlsanityData to 7 Oct 2026

Structured Neural Modeling of Daily Arctic Sea-Ice Concentration Evolution: Physical-Trajectory-Driven Learning and Forecast-Domain Adaptation

arXiv 2610.03743 ↗First posted 22 Aug 2026physics.ao-ph cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

Accurate modeling of the daily evolution of sea ice concentration (SIC) is central to improving the credibility and operational forecasting capability of deep learning-based sea ice prediction. However, existing deep learning methods often couple the underlying sea ice evolution relationships and data errors within high-dimensional nonlinear mappings, making it difficult to construct a stable and verifiable evolution core and apply it reliably to practical forecasting. To address this issue, this study proposes a reanalysis-forecast dual-domain decoupled framework for learning sea ice evolution operators. The framework builds upon a lightweight physical baseline to generate daily evolution trajectories, employs a temporally constrained joint multi-lead compensation network to com?pensate for unresolved processes, and introduces an ice-mass?aware transport mechanism to suppress numerical dissipation. In the forecasting stage, the parameters of the base evolution core are fixed, while a lightweight variable-semantic adaptation mechanism calibrates inter-domain distributions and evolution responses, thereby separating forecast-domain errors from base evolution errors. Experiments show that the constructed base evolution core can accurately and stably simulate daily sea ice evolution at both short-term and annual scales under reanal?ysis forcing, and can be effectively transferred to the forecast domain through lightweight adaptation, achieving stable prac?tical forecasting capability while preserving the base evolution structure. The source code will be made publicly available at https://github.com/zhangmaqun65535/SNM upon acceptance of this manuscript.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 26 of its 61 references that are in the map. First posted on arXiv 22 Aug 2026.

Builds on

26 of 61 references.

References · 26

  1. 2017Attention Is All You Needcited by 59,932
  2. 2015Domain-Adversarial Training of Neural Networkscited by 3,023
  3. 2017FiLM: Visual Reasoning with a General Conditioning Layercited by 2,286
  4. 2016WaveNet: A Generative Model for Raw Audiocited by 1,710
  5. 2015Convolutional LSTM Network: A Machine Learning Approach for Precipitation Nowcastingcited by 1,428
  6. 2016Deep CORAL: Correlation Alignment for Deep Domain Adaptationcited by 1,194
  7. 2015Scheduled Sampling for Sequence Prediction with Recurrent Neural Networkscited by 779
  8. 2019Temporal Fusion Transformers for Interpretable Multi-horizon Time Series Forecastingcited by 748
  9. 2017Learning multiple visual domains with residual adapterscited by 553
  10. 2023Neural General Circulation Models for Weather and Climatecited by 232
  11. 2021PredRNN: A Recurrent Neural Network for Spatiotemporal Predictive Learningcited by 181
  12. 2022SimVP: Simpler yet Better Video Predictioncited by 181
14 more
  1. 2020Integrating Scientific Knowledge with Machine Learning for Engineering and Environmental Systemscited by 160
  2. 2022Earthformer: Exploring Space-Time Transformers for Earth System Forecastingcited by 146
  3. 2018PredRNN++: Towards A Resolution of the Deep-in-Time Dilemma in Spatiotemporal Predictive Learningcited by 122
  4. 2018Neural networks for post-processing ensemble weather forecastscited by 95
  5. 2020Combining data assimilation and machine learning to infer unresolved scale parametrisationcited by 40
  6. 2020Using machine learning to correct model error in data assimilation and forecast applicationscited by 40
  7. 2024VMRNN: Integrating Vision Mamba and LSTM for Efficient and Accurate Spatiotemporal Forecastingcited by 35
  8. 2023SwinLSTM:Improving Spatiotemporal Prediction Accuracy using Swin Transformer and LSTMcited by 32
  9. 2019STConvS2S: Spatiotemporal Convolutional Sequence to Sequence Network for Weather Forecastingcited by 23
  10. 2023MT-IceNet -- A Spatial and Multi-Temporal Deep Learning Model for Arctic Sea Ice Forecastingcited by 5
  11. 2023Hierarchical Information-sharing Convolutional Neural Network for the Prediction of Arctic Sea Ice Concentration and Velocitycited by 5
  12. 2025Frequency-Compensated Network for Daily Arctic Sea Ice Concentration Predictioncited by 4
  13. 2026FloeNet: A mass-conserving global sea ice emulator that generalizes across climatescited by 4
  14. 2025Task-Adaptive Parameter-Efficient Fine-Tuning for Weather Foundation Modelscited by 3

Built on by

Not cited yet.