The application of deep learning in high‐precision ionospheric parameter prediction has become one of the focus in space weather research. In this study, an improved model called Mixed Convolutional Neural Networks (CNN)—Bi‐Long Short Term Memory is proposed for predicting future ionospheric Total Electron Content (TEC). The model is trained using the longest available (25 years) Global Ionospheric Maps‐TEC and evaluated the accuracy of ionospheric storm predictions. The results indicate that using historical TEC in the solar‐geographical reference frame as input driving data achieves higher prediction accuracy compared to that in the geocentric coordinate system. Additionally, by comparing different input parameters, it is found that incorporating the Kp, ap, and Dst indices as inputs to the model effectively improves its accuracy, especially in long‐term forecasting where R² increased by 3.49% and Root Mean Square Error decreased by 13.48%. Compared with BiLSTM‐Deep Neural Networks (DNN) and CNN‐BiLSTM, the Mixed CNN‐BiLSTM model has the highest prediction accuracy. It suggests that the utilization of CNN modules for processing spatial information, along with the incorporation of DNN modules to incorporate geomagnetic indices for result correction. Moreover, in short‐term predictions, the model accurately forecasts the evolution process of ionospheric storms. When extending the predicted length, although there are cases of prediction errors, the model still captures the entire process of ionospheric storms. Furthermore, the predicted results are significantly influenced by longitude, magnetic latitude, and local time.

TIMED/GUVI limb measurements and first‐principles simulations from the Thermosphere Ionosphere Electrodynamics Global Circulation Model (TIEGCM) are used to investigate thermospheric atomic oxygen (O) and molecular nitrogen (N₂) responses in the middle thermosphere on a constant pressure surface (∼160 km) to the November 20 and 21, 2003 superstorm. The consistency between GUVI observations and TIEGCM simulated composition changes allows us to utilize TIEGCM outputs to investigate the storm‐time behaviors of O and N₂ systematically. Diagnostic analysis shows that horizontal and vertical advection are the two main processes that determine the storm‐induced perturbations in the middle thermosphere. Molecular diffusion has a relatively smaller magnitude than the two advection processes, acting to compensate for the changes caused by the transport partly. Contributions from chemistry and eddy diffusion are negligible. During the storm initial and main phases, composition variations at high latitudes are determined by both horizontal and vertical advection. At middle‐low latitudes, horizontal advection is the main driver for the composition changes where O mass mixing ratio decreases (N₂ mass mixing ratio increases); whereas horizontal and vertical advection combined to dominate the changes in the regions where increases ( decreases). Over the entire storm period, horizontal advection plays a significant role in transporting high‐latitude composition perturbations globally. Our results also demonstrate that storm‐time temperature changes are not the direct cause of the composition perturbations on constant pressure surfaces.

TIMED/Global Ultraviolet Imager (GUVI) limb measurements of far‐ultraviolet airglow emission have been used to investigate middle‐low latitude thermospheric composition and neutral temperature responses to the 20 and 21 November 2003 (day of year [DOY] 324 and 325) superstorm. Altitude profiles of O, N₂ number densities and temperature, as well as O/N₂ column density ratio (∑O/N₂), on the storm days along the GUVI limb tracks are compared with those on DOY 323 (quiet time). The storm‐time composition and temperature responses were global and evolved continuously as the storm progressed. Specially, N₂ and temperature increased almost globally at all altitudes during the storm and their perturbation structures were similar. The magnitudes of their enhancements both increased with altitude and latitude. The storm‐induced O perturbations decreased in the lower thermosphere but increased in the upper thermosphere. Transition heights of O perturbations from decrease to increase changed with latitude and time. During the storm main and recovery phases, the storm‐induced ∑O/N₂ decreases were mostly related to the O depletion in the low‐middle thermosphere, whereas ∑O/N₂ increases during the storm were primarily caused by N₂ depletion. There was a remarkable hemispheric asymmetry in composition responses as they have different morphologies and lifetime, especially during the storm recovery phase.

An annual solar eclipse occurred on 21 June 2020. The maximum eclipse path of this solar eclipse covers the geomagnetic latitude 10–20°N in East Asia. The geomagnetic conjugate region of the eclipse region covers most of the land in Australia. It provides a valuable opportunity to study the ionospheric responses not only in the solar eclipse region but also in the conjugate region. Utilizing a coupled ionosphere and thermosphere model, we modeled global ionospheric responses and identified significant variations in electron density and electron temperature in both the eclipse region and its conjugate region. The simulated results show a significant enhancement in electron density and large decrease in electron temperature at high altitudes in the conjugate region. Two control simulations were carried out to study the main reason for the ionospheric disturbances in the conjugate hemisphere by cutting off transhemisphere photoelectron transport and thermal conduction process along field lines. The simulated results suggest that the ionospheric disturbances in the conjugate hemisphere are caused by the combined effects of transhemisphere photoelectron transport, thermal conduction, and thermospheric wind changes. Furthermore, the thermal conduction process along filed lines should be the main reason.