ST-GRIT: Spatio-Temporal Graph Transformer For Internal Ice Layer Thickness Prediction

2025-07-09

AI Summary

This study addresses a critical glaciological need: accurate spatiotemporal prediction of internal ice-layer thickness from radar imagery to support snow accumulation monitoring, ice-dynamical assessment, and reduction of climate model uncertainty. We tackle two key challenges: modeling long-range spatiotemporal dependencies across shallow and deep ice layers, and overcoming oversmoothing and poor noise robustness inherent in conventional graph neural networks (GNNs). To this end, we propose ST-GRIT—a novel architecture integrating geometric graph learning with a Graph Transformer. It features a decoupled spatiotemporal self-attention mechanism to efficiently capture long-distance dependencies on the graph structure, and incorporates local spatial feature embeddings to effectively mitigate oversmoothing and enhance noise resilience. Evaluated on real Greenland Ice Sheet radar data, ST-GRIT achieves the lowest RMSE, significantly outperforming state-of-the-art methods and diverse GNN baselines. It is the first approach to enable high-accuracy, end-to-end, joint spatiotemporal modeling of ice-layer thickness.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Reasoning under Uncertainty: Graphical Models

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Search and Retrieval-Augmented AI: Web query analysis, representation and understanding
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Understanding the thickness and variability of internal ice layers in radar imagery is crucial for monitoring snow accumulation, assessing ice dynamics, and reducing uncertainties in climate models. Radar sensors, capable of penetrating ice, provide detailed radargram images of these internal layers. In this work, we present ST-GRIT, a spatio-temporal graph transformer for ice layer thickness, designed to process these radargrams and capture the spatiotemporal relationships between shallow and deep ice layers. ST-GRIT leverages an inductive geometric graph learning framework to extract local spatial features as feature embeddings and employs a series of temporal and spatial attention blocks separately to model long-range dependencies effectively in both dimensions. Experimental evaluation on radargram data from the Greenland ice sheet demonstrates that ST-GRIT consistently outperforms current state-of-the-art methods and other baseline graph neural networks by achieving lower root mean-squared error. These results highlight the advantages of self-attention mechanisms on graphs over pure graph neural networks, including the ability to handle noise, avoid oversmoothing, and capture long-range dependencies. Moreover, the use of separate spatial and temporal attention blocks allows for distinct and robust learning of spatial relationships and temporal patterns, providing a more comprehensive and effective approach.

Problem

• Predict internal ice layer thickness from radar imagery
• Model spatiotemporal relationships in ice layer data
• Improve accuracy over existing graph neural networks

Innovation

• Spatio-temporal graph transformer for ice layers
• Inductive geometric graph learning framework
• Separate spatial and temporal attention blocks