GENERATING ELECTROMAGNETIC SATELLITE SYNTHETIC DATA FOR DETECTING SEISMIC PRECURSORS

Y Bi; Wei Zhai; Maja Pavlovic; Peter Nicholl

doi:10.1109/IGARSS55030.2025.11243082

GENERATING ELECTROMAGNETIC SATELLITE SYNTHETIC DATA FOR DETECTING SEISMIC PRECURSORS

Research output: Contribution to conference › Paper › peer-review

Abstract

The Earth’s geomagnetic field not only protects our planet
from the solar wind, but it also reflects essential evolution pro-
cesses of some physical phenomena such as earthquakes. The
literature shows that the ULF pulsations of the geomagnetic
field could contain anomalies of earthquakes and such anoma-
lies could be captured from Space. The Swarm mission con-
sists of three identical satellites which are able to measure the
geomagnetic field. In this study, we propose a deep learning
(DL) Encoder-Deconder model for generating the synthetic
Swarm data, which is used to detect seismic anomalies. This
model is built on multiple LSTM models and evaluated on
the same lengths of input and output windows. The exper-
imental results show the model’s effectiveness and the suit-
ability of DL techniques for this task, and lay a good foun-
dation for further developing effective methods for detecting
seismic anomalies from the Swarm data.

Original language	English
Pages	183-187
Number of pages	5
DOIs	https://doi.org/10.1109/IGARSS55030.2025.11243082
Publication status	Published online - 8 Aug 2025
Event	IGARSS 2025 - 2025 IEEE International Geoscience and Remote Sensing Symposium - Brisbane, Australia, Brisbane, Australia Duration: 3 Aug 2025 → 8 Aug 2025

Conference

Conference	IGARSS 2025 - 2025 IEEE International Geoscience and Remote Sensing Symposium
Abbreviated title	IGARSS 2025
Country/Territory	Australia
City	Brisbane
Period	3/08/25 → 8/08/25

Funding

European Space Agency for funding (Grant ID: 59308)

Keywords

Synthetic data generation
Deep learning
satellite data analysis
anomaly detection

Access to Document

10.1109/IGARSS55030.2025.11243082

https://www.2025.ieeeigarss.org/view_paper.php?PaperNum=4740&SessionID=1514#top

Cite this

@conference{de34b8f16e154ebe8717764d9f946c8e,

title = "GENERATING ELECTROMAGNETIC SATELLITE SYNTHETIC DATA FOR DETECTING SEISMIC PRECURSORS",

abstract = "The Earth{\textquoteright}s geomagnetic field not only protects our planet from the solar wind, but it also reflects essential evolution pro- cesses of some physical phenomena such as earthquakes. The literature shows that the ULF pulsations of the geomagnetic field could contain anomalies of earthquakes and such anoma- lies could be captured from Space. The Swarm mission con- sists of three identical satellites which are able to measure the geomagnetic field. In this study, we propose a deep learning (DL) Encoder-Deconder model for generating the synthetic Swarm data, which is used to detect seismic anomalies. This model is built on multiple LSTM models and evaluated on the same lengths of input and output windows. The exper- imental results show the model{\textquoteright}s effectiveness and the suit- ability of DL techniques for this task, and lay a good foun- dation for further developing effective methods for detecting seismic anomalies from the Swarm data.",

keywords = "Synthetic data generation, Deep learning, satellite data analysis, anomaly detection",

author = "Y Bi and Wei Zhai and Maja Pavlovic and Peter Nicholl",

year = "2025",

month = aug,

day = "8",

doi = "10.1109/IGARSS55030.2025.11243082",

language = "English",

pages = "183--187",

note = "IGARSS 2025 - 2025 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2025 ; Conference date: 03-08-2025 Through 08-08-2025",

}

TY - CONF

T1 - GENERATING ELECTROMAGNETIC SATELLITE SYNTHETIC DATA FOR DETECTING SEISMIC PRECURSORS

AU - Bi, Y

AU - Zhai, Wei

AU - Pavlovic, Maja

AU - Nicholl, Peter

PY - 2025/8/8

Y1 - 2025/8/8

N2 - The Earth’s geomagnetic field not only protects our planet from the solar wind, but it also reflects essential evolution pro- cesses of some physical phenomena such as earthquakes. The literature shows that the ULF pulsations of the geomagnetic field could contain anomalies of earthquakes and such anoma- lies could be captured from Space. The Swarm mission con- sists of three identical satellites which are able to measure the geomagnetic field. In this study, we propose a deep learning (DL) Encoder-Deconder model for generating the synthetic Swarm data, which is used to detect seismic anomalies. This model is built on multiple LSTM models and evaluated on the same lengths of input and output windows. The exper- imental results show the model’s effectiveness and the suit- ability of DL techniques for this task, and lay a good foun- dation for further developing effective methods for detecting seismic anomalies from the Swarm data.

AB - The Earth’s geomagnetic field not only protects our planet from the solar wind, but it also reflects essential evolution pro- cesses of some physical phenomena such as earthquakes. The literature shows that the ULF pulsations of the geomagnetic field could contain anomalies of earthquakes and such anoma- lies could be captured from Space. The Swarm mission con- sists of three identical satellites which are able to measure the geomagnetic field. In this study, we propose a deep learning (DL) Encoder-Deconder model for generating the synthetic Swarm data, which is used to detect seismic anomalies. This model is built on multiple LSTM models and evaluated on the same lengths of input and output windows. The exper- imental results show the model’s effectiveness and the suit- ability of DL techniques for this task, and lay a good foun- dation for further developing effective methods for detecting seismic anomalies from the Swarm data.

KW - Synthetic data generation

KW - Deep learning

KW - satellite data analysis

KW - anomaly detection

UR - https://www.2025.ieeeigarss.org/

UR - https://pure.ulster.ac.uk/en/publications/de34b8f1-6e15-4ebe-8717-764d9f946c8e

U2 - 10.1109/IGARSS55030.2025.11243082

DO - 10.1109/IGARSS55030.2025.11243082

M3 - Paper

SP - 183

EP - 187

T2 - IGARSS 2025 - 2025 IEEE International Geoscience and Remote Sensing Symposium

Y2 - 3 August 2025 through 8 August 2025

ER -

GENERATING ELECTROMAGNETIC SATELLITE SYNTHETIC DATA FOR DETECTING SEISMIC PRECURSORS

Abstract

Conference

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this