EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Graph Neural Networks with dilated convolutions for geomagnetic field forecasting: a single-station topological approach

Bryan Tipán, Ericson López, William Carvajal, Leonel Guamán

Frontiers in Astronomy and Space Sciences · 19.08.2026 · 13

DOI: 10.3389/fspas.2026.1854971 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Frontiers in Astronomy and Space Sciences, online erschienen am 19.08.2026. Verfasst von Bryan Tipán und 3 weiteren Autor:innen. Die Arbeit nennt 22 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Geomagnetic field forecasting is critical for mitigating space weather hazards, yet single-station prediction remains a challenge due to the complex, non-linear coupling of vector components. In this work, we propose a Graph Neural Network (GNN) architecture enhanced with Temporal Convolutional Networks (TCN) to forecast the H, D, and Z components. By modeling the observatory’s sensors as nodes in a learned directed graph, the system captures dynamic spatio-temporal correlations between orthogonal components. We introduce a dilated inception layer to efficiently capture multi-scale temporal patterns. Experimental results using 1-min downsampled data from the MAGDAS station (Ecuador) show that our model achieves a Mean Absolute Error (MAE) of 0.8036 nT, outperforming Vector Autoregression (VAR) and thoroughly optimized LSTM baselines. While the Naive Persistence baseline yields a marginally lower global MAE due to the statistical dominance of quiet-time periods, the proposed GNN-TCN provides vastly superior phase tracking ( R = 0.9950 ) and structural stability. Rigorous evaluation under varying space weather conditions reveals robust performance, maintaining a highly controlled error even during the top 5% most severe local geomagnetic storms. These findings, supported by power spectral density analysis, indicate that high-frequency fluctuations at a single station are dominated by instrumental and stochastic noise, making temporal filtering via downsampling essential. Ultimately, this proof-of-concept study demonstrates that utilizing a learned adjacency matrix as an adaptable structural regularization, rather than extracting a fixed physical law, provides a highly stable and effective framework for short-term geomagnetic field forecasting.

Zitieren

APA

Tipán, B., López, E., Carvajal, W., & Guamán, L. (2026). Graph Neural Networks with dilated convolutions for geomagnetic field forecasting: a single-station topological approach. Frontiers in Astronomy and Space Sciences, 13. https://doi.org/10.3389/fspas.2026.1854971

Vancouver

Tipán B, López E, Carvajal W, Guamán L. Graph Neural Networks with dilated convolutions for geomagnetic field forecasting: a single-station topological approach. Frontiers in Astronomy and Space Sciences. 2026;13. doi:10.3389/fspas.2026.1854971

BibTeX
@article{Tipn2026,
  author = {Tipán, Bryan and López, Ericson and Carvajal, William and Guamán, Leonel},
  title = {Graph Neural Networks with dilated convolutions for geomagnetic field forecasting: a single-station topological approach},
  journal = {Frontiers in Astronomy and Space Sciences},
  year = {2026},
  volume = {13},
  doi = {10.3389/fspas.2026.1854971}
}
Kostenlos starten →