Machine learning-based detection of TEC signatures related to earthquakes and tsunamis - the 2015 Illapel case study

Aug 5, 2026

Upload by Asher

Earthquakes and tsunamis can generate acoustic and gravity waves that reach the ionosphere, producing traveling ionospheric disturbances manifested as changes in ionospheric total electron content (TEC) observed via GNSS receivers. The VARION algorithm estimates real-time TEC variations, enabling a data-rich machine learning workflow for TEC perturbation detection. Using VARION-generated dsTEC/dt features from 115 GNSS stations, the study applies Random Forest and XGBoost with elevation cut-offs (15° and 25°) to labeled disturbance intervals. XGBoost with a 15° cut-off achieves F1=0.77 and precision/recall of 0.80/0.74, with a 75 s average timing difference, supporting real-time early warning integration.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages14

Reading Time35 min

AccessFree to download

Abstract

Introduction

GPS Solutions (2024) 28:106 [https://doi.org/10.1007/s10291-024-01649-z](https://doi.org/10.1007/s10291-024-01649-z) Machine learning‑based detection of TEC signatures related to earthquakes and tsunamis: the 2015 Illapel case study Federica Fuso1,2 · Laura Crocetti2 · Michela Ravanelli3 · Benedikt Soja2 Received: 26 October 2023 / Accepted: 26 March 2024 / Published online: 20 April 2024 © The Author(s) 2024 Abstract Earthquakes and tsunamis can trigger acoustic and gravity waves that could reach the ionosphere, generating electron density disturbances, known as traveling ionospheric disturbances. These perturbations can be investigated as variations in ionospheric total electron content (TEC) estimated through global navigation satellite systems (GNSS) receivers. The VARION (Variometric Approach for Real-Time Ionosphere Observation) algorithm is a well-known real-time tool for estimating TEC variations. In this context, the high amount of data allows the exploration of a VARION-based machine learning classification approach for TEC perturbation detection. For this purpose, we analyzed the 2015 Illapel earthquake and tsunami for its strength and high impact. We use the VARION-generated observations (i.e., dsTEC/dt) provided by 115 GNSS stations as input features for the machine learning algorithms, namely, Random Forest and XGBoost. We manually label time frames of TEC perturbations as the target variable. We consider two elevation cut-off time series, namely, 15° and 25°, to which we apply the classifier. XGBoost with a 15° elevation cut-off dsTEC/dt time series reaches the best performance, achieving an F1 score of 0.77, recall of 0.74, and precision of 0.80 on the test data. Furthermore, XGBoost presents an average difference between the labeled and predicted middle epochs of TEC perturbation of 75 s. Finally, the model could be seamlessly integrated into a real-time early warning system, due to its low computational time. This work demonstrates high-probability TEC signature detection by machine learning for earthquakes and tsunamis, that can be used to enhance tsunami early warning systems. Keywords GNSS Ionospheric Seismology · VARION algorithm · Machine learning · XGBoost · Tsunami early warning systems * Federica Fuso [federica.fuso@uniroma1.it](federica.fuso@uniroma1.it) Laura Crocetti [lcrocetti@ethz.ch](lcrocetti@ethz.ch) Michela Ravanelli [michela.ravanelli@uniroma1.it](michela.ravanelli@uniroma1.it) Benedikt Soja [soja@ethz.ch](soja@ethz.ch) 1 Department of Computer, Control and Management Engineering (DIAG), Sapienza University of Rome, Via Ariosto, 25, 00185 Rome, Italy 2 Institute of Geodesy and Photogrammetry, ETH Zurich, Robert-Gnehm-Weg 15, 8093 Zurich, Switzerland 3 Department of Civil, Constructional and Environmental Engineering (DICEA), Sapienza University of Rome, 00184 Rome, Italy Introduction Natural hazards such as volcanic eruptions, earthquakes, and tsunamis can perturb the ionosphere (Astafyeva 2019 ; Huang et al. 2019 ; Calais and Minster 1995 ; Peltier and Hines 1976 ; Hargreaves 1992 ; Occhipinti 2015 ; Rolland et al. 2010 ; Meng et al. 2019 ; Artru et al. 2005 ; Chouet al. 2017 ; Zettergren et al. 2017) . In detail, these events can generate acoustic and gravity waves (AGWs), that are amplified as atmosphere density decreases and can reach the ionosphere. These waves interact with the ionospheric plasma and cause ionospheric electron density disturbances, known as traveling ionospheric disturbances (TIDs; Galvanet al. 2012 ; Astafyeva 2019) . Here, we mention acoustic gravity waves generated near the epicenter (AGWepi) and internal gravity waves (IGWtsu; Occhipinti 2015). AGWepi, related to the uplift at the source, reaches the ionosphere in around 8 min, whereas IGWtsu, linked to tsunami offshore propagation, takes about 45–60 min (Lognonné et al. 2006 ; Occhipinti et al. 2011) . These perturbations are detected through variations inionospheric total

FAQs

What physical phenomenon links earthquakes and tsunamis to GNSS observations?

Seismic events can trigger acoustic and gravity waves that reach the ionosphere, causing traveling ionospheric disturbances. These disturbances appear as variations in ionospheric total electron content (TEC) measurable with GNSS receivers.

How does the study use the VARION algorithm in its machine learning pipeline?

VARION provides real-time TEC variation observations, specifically dsTEC/dt, which are used as input features for classification. Time frames of TEC perturbations are manually labeled as the target variable.

Which machine learning model and elevation cut-off produced the best detection performance?

XGBoost with a 15° elevation cut-off dsTEC/dt time series delivered the best results, reaching an F1 score of 0.77 with recall 0.74 and precision 0.80 on test data.

Aug 5, 2026

Upload by Asher