K-STEMIT: Knowledge-Informed Spatio-Temporal Efficient Multi-Branch Graph Neural Network for Subsurface Stratigraphy Thickness Estimation from Radar Data

2026-04-10

AI Summary

Existing radar inversion methods for estimating internal layer thickness in polar ice sheets are prone to noise interference and lack physical constraints, often yielding distorted spatiotemporal extrapolations. This work proposes an efficient multi-branch spatiotemporal graph neural network that integrates physics-informed priors through geometric spatial modeling, temporal convolutions, and synergistic use of atmospheric model data, complemented by an adaptive feature fusion mechanism to dynamically integrate multi-source information. By embedding physics-driven priors into a multi-branch graph architecture for the first time, the method significantly enhances the plausibility and accuracy of extrapolated results while maintaining computational efficiency. Compared to state-of-the-art approaches, it achieves a 21.01% reduction in root mean square error and consistently lower interannual relative mean absolute error, enabling high-fidelity and reliable assessment of spatiotemporal variations in snow and ice accumulation.

Technology Category

• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Machine Learning: Graph-based Machine Learning
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems

Application Category

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

Abstract

Subsurface stratigraphy contains important spatio-temporal information about accumulation, deformation, and layer formation in polar ice sheets. In particular, variations in internal ice layer thickness provide valuable constraints for snow mass balance estimation and projections of ice sheet change. Although radar sensors can capture these layered structures as depth-resolved radargrams, convolutional neural networks applied directly to radar images are often sensitive to speckle noise and acquisition artifacts. In addition, purely data-driven methods may underuse physical knowledge, leading to unrealistic thickness estimates under spatial or temporal extrapolation. To address these challenges, we develop K-STEMIT, a novel knowledge-informed, efficient, multi-branch spatio-temporal graph neural network that combines a geometric framework for spatial learning with temporal convolution to capture temporal dynamics, and incorporates physical data synchronized from the Model Atmospheric Regional physical weather model. An adaptive feature fusion strategy is employed to dynamically combine features learned from different branches. Extensive experiments have been conducted to compare K-STEMIT against current state-of-the-art methods in both knowledge-informed and non-knowledge-informed settings, as well as other existing methods. Results show that K-STEMIT consistently achieves the highest accuracy while maintaining near-optimal efficiency. Most notably, incorporating adaptive feature fusion and physical priors reduces the root mean-squared error by 21.01% with negligible additional cost compared to its conventional multi-branch variants. Additionally, our proposed K-STEMIT achieves consistently lower per-year relative MAE, enabling reliable, continuous spatiotemporal assessment of snow accumulation variability across large spatial regions.

Problem

• subsurface stratigraphy
• thickness estimation
• radar data
• spatio-temporal extrapolation
• physical knowledge

Innovation

• knowledge-informed learning
• spatio-temporal graph neural network
• adaptive feature fusion
• physical priors
• subsurface stratigraphy estimation