Structured Neural Modeling of Daily Arctic Sea-Ice Concentration Evolution: Physical-Trajectory-Driven Learning and Forecast-Domain Adaptation
Abstract
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.
Related
- Harnessing fuzzy forest intelligence: A novel mechanism modeling approach for particle matter concentration in sewage sludge–municipal solid waste co-incineration
- Evolution of firework-related barium aerosols: insights from single-particle analysis and mass concentration monitoring
- Three-Dimensional Numerical Modeling of Glacier Flow and U-Shaped Valley Evolution in Subtropical Taiwan
- Trajectory of the Landes forest (1850-2022): a carbon sink subject to changing social and ecological dynamics
- Large Seasonal Variation in Microplastic Concentration from the Truckee River, United States
BibTeX
@misc{zhang2026structured,
title = {{Structured Neural Modeling of Daily Arctic Sea-Ice Concentration Evolution: Physical-Trajectory-Driven Learning and Forecast-Domain Adaptation}},
author = {Maqun Zhang and Feng Gao and Wankun Chen and Hui Yu and Yanhai Gan and Junyu Dong},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.03743},
eprint = {2610.03743},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.03743},
}
RIS
TY - UNPB
TI - Structured Neural Modeling of Daily Arctic Sea-Ice Concentration Evolution: Physical-Trajectory-Driven Learning and Forecast-Domain Adaptation
AU - Maqun Zhang
AU - Feng Gao
AU - Wankun Chen
AU - Hui Yu
AU - Yanhai Gan
AU - Junyu Dong
PY - 2026
DA - 2026/10/01
PB - arXiv
DO - 10.48550/arxiv.2610.03743
UR - https://arxiv.org/abs/2610.03743
AB - 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.
ER -