Frequency-Enhanced Hilbert Scanning Mamba for Short-Term Arctic Sea Ice Concentration Prediction

2026-02-13

AI Summary

This study addresses the limitations of conventional Mamba models in capturing temporal dependencies and preserving boundary details for short-term Arctic sea ice concentration forecasting. To overcome these challenges, the authors propose FH-Mamba, a novel framework that integrates a 3D Hilbert scanning path to maintain spatiotemporal neighborhood structures, employs wavelet transforms to enhance high-frequency details, and introduces a hybrid shuffle attention module for adaptive fusion of sequential and frequency-domain features. As the first sea ice prediction approach to combine frequency-domain enhancement with structure-aware sequential modeling, FH-Mamba achieves state-of-the-art performance on the OSI-450a1 and AMSR2 datasets, significantly improving both temporal consistency and edge reconstruction accuracy in forecasts.

Technology Category

• Machine Learning: Time-Series/Data Streams
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Computer Vision: Diffusion Models for Vision

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

While Mamba models offer efficient sequence modeling, vanilla versions struggle with temporal correlations and boundary details in Arctic sea ice concentration (SIC) prediction. To address these limitations, we propose Frequency-enhanced Hilbert scanning Mamba Framework (FH-Mamba) for short-term Arctic SIC prediction. Specifically, we introduce a 3D Hilbert scan mechanism that traverses the 3D spatiotemporal grid along a locality-preserving path, ensuring that adjacent indices in the flattened sequence correspond to neighboring voxels in both spatial and temporal dimensions. Additionally, we incorporate wavelet transform to amplify high-frequency details and we also design a Hybrid Shuffle Attention module to adaptively aggregate sequence and frequency features. Experiments conducted on the OSI-450a1 and AMSR2 datasets demonstrate that our FH-Mamba achieves superior prediction performance compared with state-of-the-art baselines. The results confirm the effectiveness of Hilbert scanning and frequency-aware attention in improving both temporal consistency and edge reconstruction for Arctic SIC forecasting. Our codes are publicly available at https://github.com/oucailab/FH-Mamba.

Problem

• Arctic sea ice concentration
• temporal correlation
• boundary details
• short-term prediction
• sequence modeling

Innovation

• Hilbert scanning
• frequency enhancement
• Mamba
• wavelet transform
• spatiotemporal modeling