EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

CNN‐Based Model for 3D CME Parameter Prediction – A Proof of Concept

Harshita Gandhi, Huw Morgan, Cory Thomas

Journal of Geophysical Research: Machine Learning and Computation · 2025-12 · 2 (4)

DOI: 10.1029/2025jh000858 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Geophysical Research: Machine Learning and Computation, online erschienen am 26.09.2025. Verfasst von Harshita Gandhi und 2 weiteren Autor:innen. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 99 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Science and Technology Facilities Council.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Accurate 3D reconstruction of coronal mass ejections (CMEs) is essential for understanding their propagation and improving space weather forecasts. In this study, we present a proof‐of‐concept deep learning framework that predicts seven parameters from synthetic multi‐view coronagraph images. We generate 9250 synthetic CME events using a 3D wireframe model, each defined by randomly sampled values for CME's longitude, latitude, tilt‐angle, and 3D propagation speed. The speed is used to compute heights at three timepoints aligned with SOHO/LASCO C3 and STEREO‐A/COR2 and STEREO‐B/COR2 cadences. These values are used to render time‐resolved synthetic images, serving as inputs to a convolutional neural network (CNN). The CNN is trained using five‐fold cross‐validation to predict the sine and cosine of the longitude, latitude, tilt‐angle, and apex heights at the three specified times. CNN achieves mean absolute errors (MAE) of 4.88°, 2.50°, and 13.35° for longitude, latitude, and tilt, and for heights. The predicted heights are used to derive CME speed via height‐time fitting. Performance metrics include Pearson's for most parameters (tilt: ), and values of 0.99 for longitude and latitude, with heights , and 0.61 for tilt. The average test MAE across all outputs is 0.134 0.004. When applied to a real CME event, the CNN's predictions fall within the typical uncertainties reported for standard geometric fitting approaches. While challenges remain in handling real data, our results indicate that the trained model can generalize beyond synthetic data, highlighting its potential for operational space weather forecasting.

Förderung

  • Science and Technology Facilities Council

Zitieren

APA

Gandhi, H., Morgan, H., & Thomas, C. (2025). CNN‐Based Model for 3D CME Parameter Prediction – A Proof of Concept. Journal of Geophysical Research: Machine Learning and Computation, 2(4). https://doi.org/10.1029/2025jh000858

Vancouver

Gandhi H, Morgan H, Thomas C. CNN‐Based Model for 3D CME Parameter Prediction – A Proof of Concept. Journal of Geophysical Research: Machine Learning and Computation. 2025;2(4). doi:10.1029/2025jh000858

BibTeX
@article{Gandhi2025,
  author = {Gandhi, Harshita and Morgan, Huw and Thomas, Cory},
  title = {CNN‐Based Model for 3D CME Parameter Prediction – A Proof of Concept},
  journal = {Journal of Geophysical Research: Machine Learning and Computation},
  year = {2025},
  volume = {2},
  number = {4},
  doi = {10.1029/2025jh000858}
}
Kostenlos starten →