An increasing number of terrestrial‐ and space‐based radio‐communication systems are influenced by the ionospheric space weather, making the ionospheric state increasingly important to forecast. In this study, a novel extended encoder‐decoder long short‐term memory extended (ED‐LSTME) neural network, which can predict ionospheric total electron content (TEC) is proposed. Useful inherent features were automatically extracted from the historical TEC by LSTM layers, and the performance of the proposed model was enhanced by considering solar flux and geomagnetic activity data. The proposed ED‐LSTME model was validated using 15‐min TEC values from GPS measurements over one solar cycle (from January 2006 to July 2018) collected at 15 GPS stations in China. Different assessment experiments were conducted in different geographical locations and seasons as well as under varying geomagnetic activities, to comprehensively evaluate the model's performance. These comparative experiments were conducted using an ED‐LSTM, a traditional LSTM, a deep neural network, autoregressive integrated moving average, and the 2016 International Reference Ionosphere models. The results indicated that the ED‐LSTME model is superior to the other statistical models, with R² and root mean square error values of 0.89 and 12.09 TECU, respectively. In addition, TEC was adequately predicted under different ionospheric conditions, and satisfactory results were obtained even under geomagnetically disturbed conditions. These results suggest that the prediction performance could be significantly improved by utilizing auxiliary data. These observations confirm that the proposed model outperforms several state‐of‐the‐art models in making predictions at different times and under diverse conditions.

Nowadays, the possibility that medium-large earthquakes could produce some electromagnetic ionospheric disturbances during their preparatory phase is controversial in the scientific community. Some previous works using satellite data from DEMETER, Swarm and, recently, CSES provided several pieces of evidence supporting the existence of such precursory phenomena in terms of single case studies and statical analyses. In this work, we applied a Worldwide Statistical Correlation approach to M5.5+ shallow earthquakes using the first 8 years of Swarm (i.e., from November 2013 to November 2021) magnetic field and electron density signals in order to improve the significance of previous statistical studies and provide some new results on how earthquake features could influence ionospheric electromagnetic disturbances. We implemented new methodologies based on the hypothesis that the anticipation time of anomalies of larger earthquakes is usually longer than that of anomalies of smaller magnitude. We also considered the signal’s frequency to introduce a new identification criterion for the anomalies. We find that taking into account the frequency can improve the statistical significance (up to 25% for magnetic data and up to 100% for electron density). Furthermore, we noted that the frequency of the Swarm magnetic field signal of possible precursor anomalies seems to slightly increase as the earthquake is approaching. Finally, we checked a possible relationship between the frequency of the detected anomalies and earthquake features. The earthquake focal mechanism seems to have a low or null influence on the frequency of the detected anomalies, while the epicenter location appears to play an important role. In fact, land earthquakes are more likely to be preceded by slower (lower frequency) magnetic field signals, whereas sea seismic events show a higher probability of being preceded by faster (higher frequency) magnetic field signals.

In this paper we provide a comprehensive comparison of in situ electron density (Ne) and temperature (Te) measured by Langmuir probe (LAP) on board the China Seismo‐Electromagnetic Satellite (CSES), with nearly simultaneous measurements from the Swarm B satellite, incoherent scatter radar (ISR) at Millstone Hill, as well as predictions from empirical models including the incoherent scatter radar ionospheric model (ISRIM) for Millstone Hill and International Reference Ionosphere model (IRI‐2016). Results reveal that the global distributions and their relative variations of Ne/Te from CSES and Swarm are quite consistent during conjunction periods of the two satellites, although the absolute values of Swarm Ne are proportionally larger than that of CSES. The large‐scale ionospheric structures, such as the equatorial ionization anomaly (EIA), the longitudinal wave number (WN3/4), the Weddell Sea anomaly (WSA), the northern midlatitude summer nighttime anomaly (MSNA), and the midlatitude ionospheric trough (MIT), are well represented by the CSES measurements. For the temporal variation over Millstone Hill station, CSES Ne at nightside shows some different characteristics from the predictions of IRI and ISRIM, possibly due to the influences of MIT and midlatitude arc (MLA) that are often observed at latitudes of Millstone Hill. Our results suggest that the CSES in situ plasma parameters are reliable with a high scientific potential for investigation of geophysics and space physics.

Oppositely to a previous statistical work using a single time resolution of the total ion density measured onboard the Detection of Electro‐Magnetic Emissions Transmitted from Earthquake Regions (DEMETER) satellite, this work deals with statistical seismo‐ionospheric influences by comparing different parameters and various time resolutions. The O⁺ density and electron density recorded by the China Seismo‐Electromagnetic Satellite (CSES) for more than one year and by the DEMETER satellite for about 6.5 years have been utilized to globally search ionospheric perturbations with different time resolutions. A comparison is automatically done by software between the occurrence of these ionospheric perturbations determined by different data sets, and the occurrence of earthquakes under the conditions that these perturbations occur at less than 1,500 km and up to 15 days before the earthquakes. Combined with statistical results given by both satellites, it is shown that the detection rate r of earthquakes increases as the data time resolution and the earthquake magnitude increase and as the focal depth decreases. On average, the number of perturbations is higher the day of the earthquake, and then smoothly decreases the days before, which is independent of either ionospheric parameters or time resolutions. The number of right alarms is high near the South Atlantic Magnetic Anomaly area but its relationship with seismic activities is weak. The ion density tends to be more sensitive to seismic activities than the electron density but this needs further investigations. This study shows that the CSES could effectively register ionospheric perturbations due to strong EQs as the DEMETER satellite does.