EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Deep Learning Forecasting of IRI-2020 Total Electron Content Residuals over China and Adjacent Regions During the Solar Cycle 25 Maximum: Roles of Anchored Output and the Background Field

Ren Wang, Ying Li, Qiyong Zhang, Xiaoming Li, Tingchen Jiang, Xiangyu Tian

Atmosphere · 20.11.2026 · 17 (11) · 1151

DOI: 10.3390/atmos17111151 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Atmosphere, online erschienen am 20.11.2026. Verfasst von Ren Wang und 5 weiteren Autor:innen. Die Arbeit nennt 51 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, the Basic Research Program of Jiangsu, the 2025 Lianyungang Key Research and Development Program.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Deep learning has rarely been used to forecast corrections of the International Reference Ionosphere (IRI) total electron content (TEC). We forecast the difference between the Center for Orbit Determination in Europe (CODE) global ionospheric maps and IRI-2020 TEC (the residual) over China and adjacent regions at 1–24 h lead times with four deep neural networks, trained on 2008–2019, validated on 2020–2022 and tested during the Solar Cycle 25 maximum (2023–2025). Reusing the residual observed one day earlier lowers the root mean square error (RMSE) of IRI by 37.3%. At 24 h, direct-output networks do not significantly beat this naive forecast, whereas all four networks anchored to it, predicting only the day-to-day change, beat it significantly—the patch time series Transformer (PatchTST) by 12.4%. IRI then contributes only its day-to-day increment, which helps only if the regression coefficient β of the observed TEC change onto this increment exceeds 1/2. With IRI driven by indices observed at the target time, β is 0.23 in quiet periods but 0.90 for disturbance storm time (Dst) index ≤ −50 nT. Without IRI, anchored PatchTST beats the CODE 1-day predicted maps (C1PG) by 11.1%, a comparison favored by its final-map inputs. Naive observational forecasts should be routine benchmarks; β indicates when a background field adds predictive value.

Förderung

  • National Natural Science Foundation of China
  • the Basic Research Program of Jiangsu
  • the 2025 Lianyungang Key Research and Development Program
  • the 2025 University-level Undergraduate Education and Teaching Reform Project of Jiangsu Ocean University
  • the Chuzhou University Research Initiation Fund Project

Zitieren

APA

Wang, R., Li, Y., Zhang, Q., Li, X., Jiang, T., & Tian, X. (2026). Deep Learning Forecasting of IRI-2020 Total Electron Content Residuals over China and Adjacent Regions During the Solar Cycle 25 Maximum: Roles of Anchored Output and the Background Field. Atmosphere, 17(11), 1151. https://doi.org/10.3390/atmos17111151

Vancouver

Wang R, Li Y, Zhang Q, Li X, Jiang T, Tian X. Deep Learning Forecasting of IRI-2020 Total Electron Content Residuals over China and Adjacent Regions During the Solar Cycle 25 Maximum: Roles of Anchored Output and the Background Field. Atmosphere. 2026;17(11):1151. doi:10.3390/atmos17111151

BibTeX
@article{Wang2026,
  author = {Wang, Ren and Li, Ying and Zhang, Qiyong and Li, Xiaoming and Jiang, Tingchen and Tian, Xiangyu},
  title = {Deep Learning Forecasting of IRI-2020 Total Electron Content Residuals over China and Adjacent Regions During the Solar Cycle 25 Maximum: Roles of Anchored Output and the Background Field},
  journal = {Atmosphere},
  year = {2026},
  volume = {17},
  number = {11},
  pages = {1151},
  doi = {10.3390/atmos17111151}
}
Kostenlos starten →