Attention-Based Reconstruction of Full-Field Tsunami Waves from Sparse Tsunameter Networks
Machine Learning
2025-07-22 v5 Fluid Dynamics
Abstract
We investigate the potential of an attention-based neural network architecture, the Senseiver, for sparse sensing in tsunami forecasting. Specifically, we focus on the Tsunami Data Assimilation Method, which generates forecasts from tsunameter networks. Our model is used to reconstruct high-resolution tsunami wavefields from extremely sparse observations, including cases where the tsunami epicenters are not represented in the training set. Furthermore, we demonstrate that our approach significantly outperforms the Linear Interpolation with Huygens-Fresnel Principle in generating dense observation networks, achieving markedly improved accuracy.
Cite
@article{arxiv.2411.12948,
title = {Attention-Based Reconstruction of Full-Field Tsunami Waves from Sparse Tsunameter Networks},
author = {Edward McDugald and Arvind Mohan and Darren Engwirda and Agnese Marcato and Javier Santos},
journal= {arXiv preprint arXiv:2411.12948},
year = {2025}
}