Comparative analysis of empirical and deep learning models for ionospheric sporadic E layer prediction

• Published in: Earth and planetary physics Vol. 9; no. 1; pp. 10 - 19
• Main Authors: Yu, BingKun, Tian, PengHao, Xue, XiangHui, J. Scott, Christopher, Ye, HaiLun, Wu, JianFei, Yi, Wen, Chen, TingDi, Dou, XianKang
• Format: Journal Article
• Language: English
• Published: Institute of Deep Space Sciences,Deep Space Exploration Laboratory,Hefei 230088,China%CAS Key Laboratory of Geospace Environment,School of Earth and Space Sciences,University of Science and Technology of China,Hefei 230026,China%CAS Key Laboratory of Geospace Environment,School of Earth and Space Sciences,University of Science and Technology of China,Hefei 230026,China 2025; Hefei National Laboratory,University of Science and Technology of China,Hefei 230088,China%Department of Meteorology,University of Reading,Reading RG6 6BB,UK%Hefei National Laboratory,University of Science and Technology of China,Hefei 230088,China; Science Press
• Subjects: artificial intelligence; deep learning model; ionosondes; ionospheric sporadic e layer; numerical model; radio occultation; deep learning model; ionosondes; ionospheric sporadic E layer; numerical model; radio occultation; artificial intelligence
• ISSN: 2096-3955; 2096-3955
• DOI: 10.26464/epp2024048

Sporadic E(Es)layers in the ionosphere are characterized by intense plasma irregularities in the E region at altitudes of 90-130 km.Because they can significantly influence radio communications and navigation systems,accurate forecasting of Es layers is crucial for ensuring the precision and dependability of navigation satellite systems.In this study,we present Es predictions made by an empirical model and by a deep learning model,and analyze their differences comprehensively by comparing the model predictions to satellite RO measurements and ground-based ionosonde observations.The deep learning model exhibited significantly better performance,as indicated by its high coefficient of correlation(r=0.87)with RO observations and predictions,than did the empirical model(r=0.53).This study highlights the importance of integrating artificial intelligence technology into ionosphere modelling generally,and into predicting Es layer occurrences and characteristics,in particular.