Work

Attention‐Based Reconstruction of Full‐Field Tsunami Waves From Sparse Tsunameter Networks

Year: 2025

Type: article

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.

Source: Geophysical Research Letters

DOI: 10.1029/2025gl115345

Authors: Edward McDugald, Arvind Mohan, Darren Engwirda, Agnese Marcato, Javier E. Santos

Institutions: Los Alamos National Laboratory, University of Arizona, Commonwealth Scientific and Industrial Research Organisation

Language: English

Topic: Earthquake and Tsunami Effects

Subfield: Civil and Structural Engineering

Field: Engineering

Domain: Physical Sciences

Open Access status: gold

Estimated APC paid: $3,190

Funders: Laboratory Directed Research and Development, Los Alamos National Laboratory