Advancing ionospheric irregularity reconstruction with ICON/MIGHTI wind‐driven insights

Tian, P. ORCID: https://orcid.org/0000-0001-8976-2711, Yu, B. ORCID: https://orcid.org/0000-0003-2758-1960, Xue, X. ORCID: https://orcid.org/0000-0002-4541-9900, Yamazaki, Y. ORCID: https://orcid.org/0000-0002-7624-4752, Owens, M. J. ORCID: https://orcid.org/0000-0003-2061-2453, Ye, H. ORCID: https://orcid.org/0000-0002-8371-9760, Wu, J. ORCID: https://orcid.org/0000-0002-3998-9296, Chen, T. ORCID: https://orcid.org/0000-0002-8399-4084, Scott, C. J. ORCID: https://orcid.org/0000-0001-6411-5649 and Dou, X. ORCID: https://orcid.org/0000-0001-6433-6222 (2025) Advancing ionospheric irregularity reconstruction with ICON/MIGHTI wind‐driven insights. Geophysical Research Letters, 52 (12). e2025GL115666. ISSN 0094-8276 doi: 10.1029/2025GL115666

Abstract/Summary

In the mesosphere‐lower thermosphere region, atmospheric plasma components exhibit short‐term enhancements, forming sporadic E (Es) layers that impact communication systems. The prevailing theory posits that neutral wind shear is the primary driver of mid‐latitude Es layer. Here, we present neutral wind field data from the ICON/Michelson Interferometer for Global High‐resolution Thermospheric Imaging mission during 2019–2022, revealing a clear relationship between wind shear and Es layer formation in the Northern Hemisphere. Notably, the vertical ion divergence/convergence significantly impact mid‐latitude Es production. Inspired by deep learning techniques, we developed a deep learning model based on wind shear and neutral wind data, reconstructing the small‐scale morphology of Es layers. Vertical ion convergence information derived from the wind shear physical equations was found to be a key factor in enhancing model performance. Our results demonstrate that incorporating physical data from vertical ion drift improves the predictive capabilities of ionospheric irregularities artificial intelligence models, increasing the accuracy from 71.6% to 87.9%

Item Type	Article
URI	https://centaur.reading.ac.uk/id/eprint/123269
Identification Number/DOI	10.1029/2025GL115666
Refereed	Yes
Divisions	Science > School of Mathematical, Physical and Computational Sciences > Department of Meteorology
Publisher	American Geophysical Union
Download/View statistics	View download statistics for this item