Attention-Based Reconstruction of Full-Field Tsunami Waves from Sparse Tsunameter Networks

2024-11-20

AI Summary

To address the challenge of reconstructing full-field tsunami wavefields from sparse buoy observations in early-warning systems, this paper proposes Senseiver—a novel end-to-end, physics-guided neural network incorporating attention mechanisms. It is the first to integrate attention into tsunami data assimilation, enabling high-fidelity and strongly generalizable wavefield reconstruction for unseen seismic sources via sparse spatiotemporal observation modeling and a source-agnostic training paradigm. Compared to conventional linear interpolation augmented with the Huygens–Fresnel principle, Senseiver achieves significantly higher reconstruction accuracy—particularly under extremely sparse observational configurations (e.g., only 3–5 buoys) and unknown source scenarios—while maintaining robust cross-source generalization. This work overcomes the dual limitations of existing methods in both reconstruction fidelity and adaptability across diverse earthquake sources.

Technology Category

• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Computer Vision: Remote Sensing / Geospatial AI
• Humans and AI: Brain-Sensing and Analysis

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Systems and Infrastructure for Web, Mobile and WoT: Experiences and lessons learnt from Web-based algorithms and system deployments

Abstract

We investigate the potential of an attention-based neural network architecture known as the Senseiver to perform sparse sensing tasks in the context of tsunami forecasting. In particular, we focus on the Tsunami Data Assimilation Method, where forecasts are derived from tsunameter networks. We used our model to generate high-resolution tsunami waves from incredibly sparse observations, whose epicenters are not included in the training set. We also show significantly improved accuracy in the generation of dense observation networks by comparison to the Linear Interpolation with Huygens-Fresnel Principle.

Problem

• Reconstruct tsunami waves using sparse data
• Enhance tsunami forecasting accuracy
• Apply attention-based neural networks

Innovation

• Attention-based neural network
• Sparse tsunami wave reconstruction
• Improved accuracy dense networks