The Constellation Observing System for Meteorology, Ionosphere, and Climate (COSMIC) provides global Radio Occultation (RO) measurements of ionospheric total electron content (TEC), but these values are systematically underestimated relative to ground-based Global Navigation Satellite System (GNSS)-derived TEC due to the exclusion of the plasmaspheric contribution. This study presents a machine learning calibration framework that transforms COSMIC TEC into GNSS-equivalent values. Using co-located COSMIC and GNSS observations from 2006 to 2025, we developed neural network models (ROTEC-A and ROTEC-B) trained on (19 and 22) input features respectively, including COSMIC profile parameters, spatiotemporal descriptors, and optionally, solar and geomagnetic activity indices. Results show that the calibration effectively mitigates systematic underestimation, reducing mean bias from 6.97 TECU (uncalibrated COSMIC) to near zero (0.02-0.03 TECU). The calibrated products also substantially reduce skewness in residuals, yielding nearly symmetric error distributions suitable for data assimilation. Across various latitudinal, local time, and seasonal sectors, mean absolute errors were reduced by 50-75%, with the best performance at mid-latitudes and slightly elevated errors in high-latitude and equatorial regions. Although, the inclusion of solar and geomagnetic indices yielded marginal improvements, statistical tests confirmed no significant advantage over the baseline model. The operationally oriented framework outputs calibrated GNSS-equivalent TEC in near real-time, providing enhanced ionospheric monitoring capability, especially over GNSS-sparse regions such as oceans and deserts.These results demonstrate the potential of COSMIC RO data, once calibrated, to serve as a reliable complement to GNSS observations for ionospheric research, space weather monitoring, and operational applications. (c) 2026 COSPAR. Published by Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies.

The ionosphere is one of the most important layers of the atmosphere, and for its electric properties is used in communication and navigation services. In addition to being influenced by geomagnetic and solar activity, in recent decades, it has been observed that the ionosphere may also exhibit variability due to effects caused by events of terrestrial origin, such as earthquakes. However, objectively identifying when a variation is an anomaly related to an earthquake remains a challenge.This study presents a methodology based on machine learning to automatically detect the relationship between this type of irregularity and earthquakes. For this purpose, electron density (Ne) data recorded by the European Space Agency’s Swarm satellite constellation are used. Following the previously published NeAD anomaly detection algorithm, a combination of machine learning techniques is applied to group the detected anomalies according to their characteristics, to correct and automatically distinguishing the anomalies truly associated with the earthquake under study.As a case study, the Mw 7.6 earthquake that occurred in Mexico on September 19, 2022, is presented. Five types of anomalies were distinguished, showing that duration and intensity are the most important factors for differentiating them.The results suggest that one of the five anomaly groups can be associated exclusively with processes related to the main earthquake, while the other four groups are linked to other phenomena such as other minor earthquakes, tropical cyclones, or volcanic eruptions. This automated approach opens new possibilities for improving the classification of ionospheric anomalies and understanding how the lithosphere, atmosphere, and ionosphere interact with each other in the dynamics of our planet.

This study compared the ionospheric irregularities as observed using two different techniques, namely; the Constellation Observing System for Meteorology, Ionosphere, and Climate (COSMIC) satellites and the scintillation intensity index (S4) data measured by the Scintillation Network and Decision Aid (SCINDA) receiver which operated at Nairobi University (geog lon 36.8 degrees E, geog lat 1.3 degrees S, dip lat-24.1 degrees), Kenya. The data compared were those of the years 2009 (low solar activity) and 2011 (ascending phase of solar cycle 24), for both quiet (Kp < 3) and disturbed (Kp >= 5) geomagnetic conditions. For the cases of Global Positioning System (GPS) satellites with elevation angle 0 degrees as observed by the COSMIC satellites, a geo-location of the COSMIC S4 data associated with the link between GPS and COSMIC satellites was proposed at the tangent point. The COSMIC S4 data whose geo-locations fall in the vicinity of Nairobi were compared with the S4 data measured by the SCINDA receiver. The coefficient of determination which represents the percentage of the variation in COSMIC S4 data associated with the variation in SCINDA S4 was 50 %. The two data sets depict that scintillation occurs mostly in the seasons of March and September equinoxes of high solar activity conditions. However, there was a moderate positive correlation (Pearson correlation coefficient, r = 0.52 on quiet days) between COSMIC and SCINDA S4 data. The results presented signify that the COSMIC S4 could be analyzed to study ionospheric irregularities (which cause scintillations) over locations such as deserts and oceans where it is usually difficult to deploy equipment. (c) 2026 COSPAR. Published by Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies.

This study provides a multifractal assessment of ionospheric dynamics by analyzing temporal and scale-dependent variations during two distinct phases of the solar cycle: solar maximum (2014) and solar minimum (2019). Fine-temporal-resolution datasets of vertical Total Electron Content (vTEC), the SYM-H and AE geomagnetic indices, as well as solar EUV and X-ray fluxes, are analyzed using structure function (SF) and multifractal detrended fluctuation analysis (MFDFA). The analysis is conducted at a mid-latitude ionospheric station, taking into account that ionospheric dynamics show strong latitudinal dependence, with more pronounced inhomogeneities typically observed in equatorial and high-latitude regions. The SF scaling exponents reveal deviations from classical Kolmogorov scaling, particularly in 2014, indicating intermittent and nonlinear dynamics driven by enhanced solar and geomagnetic activity. MFDFA results show that vTEC and X-ray flux exhibit the broadest singularity spectra and strongest multifractality in 2014, associated with flare-related bursts and intensified ionospheric variability. In contrast, the narrower spectra observed in 2019 reflect reduced solar forcing and more uniform system dynamics. Comparisons between original and shuffled datasets demonstrate that temporal correlations rather than amplitude distributions are the dominant source of multifractality, especially in vTEC, AE, and SYM-H. These findings highlight how solar cycle variability governs the multifractal nature of ionospheric and heliospheric parameters, emphasizing the value of scale-dependent analysis for advancing space weather research. (c) 2025 COSPAR. Published by Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies.

Equatorial plasma bubbles (EPBs) disrupt satellite-based communication and navigation systems, particularly in equatorial regions. Reliable detection and classification of EPBs from all-sky imager (ASI) images are essential for accurate space weather monitoring and forecasting. This study presents a novel bootstrapping convolutional neural network (CNN) approach to optimize automated EPB detection on ASI images for operational space weather monitoring applications, and overcoming challenges related to image variability and imbalanced data sets. Data used for CNN training were obtained from the optical mesosphere thermosphere imagers ASI installed at the Space Environment Research Laboratory, National Space Research and Development Agency, Abuja during the period from 2015 to 2020. Our method involved training three sub-models, and aggregating their predictions. The CNN trainings were conducted on three sub-datasets of 3,000 images each, categorized as "EPB," "Noisy/Cloudy" or "No EPB." Three corresponding sub-models were developed from the CNN trainings. The three sub-model classifications independently gave prediction accuracies of 98.67%, 98.33%, and 95.83% on a reserved test data set of 600 images. Ensemble models further improved the model prediction accuracies to 99.17% and 99.33% for methods based on the mean of sub-model probabilities and the mode of sub-model classifications respectively. Our results indicate that the bootstrapping CNN technique enhanced the EPB detection accuracy, providing a powerful tool for real-time space weather monitoring applications, and implications for improving operational reliability of satellite-based navigation and communication in the equatorial region.

This study examines the ingestion of total electron content (TEC) data from a network of ground‐based Global Navigation Satellite System (GNSS) receivers into the NeQuick 2 model, providing a benchmark for evaluating the cost‐effectiveness of GNSS data ingestion for near‐real‐time ionospheric specification. A significant reduction in root mean square error (RMSE) between NeQuick 2 outputs and GNSS‐derived TEC was observed across various test stations, including those located at mid‐latitudes. During geomagnetically quiet periods, the performance difference between using 35 and 166 GNSS stations was less than 1%, while a similar trend, less than 2% difference was observed during disturbed periods with 34 and 177 stations, respectively. These findings highlight the importance of the spatial distribution of GNSS receivers in enhancing the model's accuracy, particularly over regions with complex ionospheric dynamics. Diurnal variations in vertical TEC (vTEC) over the test locations showed remarkable improvement during geomagnetic disturbances, with enhancements ranging from 37% to 68% in the low‐latitude region. Additionally, comparisons between NeQuick 2‐derived plasma frequencies and digisonde observations demonstrated strong agreement in the 100–200 km altitude range, with average correlation coefficients of 0.97 and 0.95 at São Luís (2.58°S, 44.20°W), 0.93 and 0.55 at Boa Vista (2.83°N, 60.70°W), and 0.88 for both conditions at Campo Grande (20.40°S, 54.50°W) during quiet and disturbed periods, respectively. The lower correlation observed at Boa Vista during disturbed conditions may be attributed to limited data ingestion coverage in the Northern Hemisphere.

The Swarm constellation is a triplet of satellites, flying, since their final configuration reached in April 2014, at altitudes of about 420 to 490 km (the lower pair) and about 500 to 530 km (the upper satellite). All the three satellites provide in-situ measurements of the plasma density in the topside ionosphere using Langmuir Probe sensors onboard the Electrical Field Instrument. The present study is a comprehensive investigation into the climatologic performance of three ionospheric models when compared to the Swarm satellite in-situ measurements. The models are the International Reference Ionosphere (IRI) model, a quick run ionospheric electron density model (NeQuick), and a 3-dimensional electron density model based on artificial neural network training of COSMIC (Constellation Observing System for Meteorology, Ionosphere, and Climate) satellites radio occultation measurements (3D-NN). The mean monthly quiet-time latitudinal profile of Swarm measurements was computed by binning the Swarm electron density measurements in 15- degree longitudes starting from longitude-180 degrees in steps of 15 degrees to 180 degrees, and corresponding model predictions were obtained. The data used in the study covers the years 2014, 2016, 2019, and 2022, capturing various phases of the solar activity cycle. Results from the study show that modelled electron density predictions from all three climatologic models are fairly good representations of the Swarm satellite measurements, with some exceptions in which the models underestimate or overestimate the Swarm satellite values. The IRI model performed best at the northern hemisphere mid latitude, and it overestimated the Swarm measurements at altitudes of ti 450 km, especially at the southern hemisphere mid and high latitudes. The NeQuick performed best during the night times, and it overestimated the Swarm measurements, especially at the mid latitudes. The NeQuick was also observed to overestimate the Swarm measurements during the winter solstices at both hemispheres, which is June solstice in the southern hemisphere and December solstice in the northern hemisphere. Overall, the 3D-NN model most often performed better than the IRI model and the NeQuick, especially during the day times and during the high solar activity year (2014), but it underestimated the Swarm measurements, especially at the low and mid latitudes. For all categories explored in the study, the 3D-NN consistently performed better than the other two models. The NeQuick performed better than the IRI model at altitude of satellite B, while the IRI model performed slightly better than the NeQuick at altitude of satellites A and C. The NeQuick also performed better than the IRI model in the local time category, whereas the IRI model performed better than the NeQuick in categories of season, solar activity, and longitudinal sector. (c) 2024 COSPAR. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/ by-nc-nd/4.0/).

This study analyzes the ionospheric dynamics during the solar maximum of 2014 and the solar minimum of 2019, focusing on Vertical Total Electron Content (vTEC) and key solar and geomagnetic indices, including SYM-H, X-ray flux, and Extreme Ultraviolet (EUV) irradiance. By employing the Hurst exponent and Probability Density Function (PDF) analysis, we quantify the persistence and correlation properties of ionospheric fluctuations under varying solar conditions. The Hurst exponent reveals significant long-range correlations in vTEC, indicating a high level of persistence, particularly during solar minimum. In contrast, solar maximum conditions exhibit more unstable behavior across all indices, with lower Hurst values suggesting enhanced short-term irregularities. PDF analysis shows leptokurtic distributions, highlighting the prevalence of extreme events, especially during heightened solar activity. Our findings underscore the complex interplay between solar activity and ionospheric behavior, providing valuable insights for improving predictive models related to space weather impacts on communication and navigation systems. (c) 2025 COSPAR. Published by Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies.

Radio Occultation (RO) is a very powerful technique to probe a planetary atmosphere, in providing vertical density profiles of the neutral atmosphere and ionosphere. The standard method uses a radio link between a spacecraft and an Earth ground station. Nevertheless, the possibility to obtain information about the Martian atmosphere with mutual RO events, using data from NASA Mars Odyssey and Mars Reconnaissance Orbiters (MRO), has been demonstrated by Ao et al. (2015).Taking advantage of two European spacecraft in orbit around Mars, the European Space Agency is currently preparing experiments of mutual RO between Mars Express (MEX) and the ExoMars Trace Gas Orbiter (TGO). In preparation of MEX and TGO data inversion and analysis, a simulation-based strategy has been adopted and an algorithm able to retrieve vertical electron density profiles from Doppler shift measurements has been implemented and validated. Subsequently, in order to test the mentioned algorithm with experimental data, the same three RO events considered in the paper by Ao et al. (2015) have been processed. In particular, for each RO event, having the information about the satellites’ orbit, the (excess) Doppler shift values corresponding to the Mars Odyssey-MRO ray-paths have been converted to bending angles as a function of impact parameter. Then, assuming a spherical symmetry (Fjeldbo et al., 1971) for the ionosphere electron density, the bending angles have been transformed (through Abel integral) to a vertical refractivity profile, which, in turn, has been converted to an ionospheric electron density profile.In this work, the results obtained by the application of the mentioned inversion algorithm to experimental data will be presented, with particular focus on the retrieval of the ionospheric electron density profiles.ReferencesAo, C. O., C. D. Edwards Jr., D. S. Kahan, X. Pi, S. W. Asmar, and A. J. Mannucci (2015), A first demonstration of Mars crosslink occultation measurements, Radio Sci., 50, 997–1007, doi:10.1002/2015RS005750.Fjeldbo, G., A. J. Kliore, and V. R. Eshleman (1971), The neutral atmosphere of Venus as studied with the Mariner V radio occultationexperiments, Astron. J., 76, 123–140.

In this paper, we investigate and propose the application of an unsupervised machine learning clustering method to characterize the spatial and temporal distribution of ionospheric plasma irregularities over the Western African equatorial region. The ordinary Kriging algorithm was used to interpolate the rate of change of the total electron content (TEC) index (ROTI) over gridded 0.5° by 0.5° latitude and longitude regional maps in order to simulate the level of ionospheric plasma irregularities in a quasi-real-time scenario. K-means was used to obtain a spatial mean index through an optimal stratification of regional post-processed ROTI maps. The results obtained could be adapted by appropriate K-means algorithms to a real-time scenario, as has been performed for other applications. This method could allow us to monitor plasma irregularities in real time over the African region and, therefore, lead to the possibility of mitigating their effects on satellite-based location systems in the said region.

Radio Occultation is a very powerful technique to probe a planetary atmosphere, in providing vertical density profiles of the neutral atmosphere and ionosphere. The standard method uses a radio link at S and/or X band between a spacecraft and an Earth ground station. At Mars, such measurements are conducted since the 60s. The three most recent data sets are from MGS (1998-2006), Mars Express (since 2004) and MAVEN (since 2016). Taking advantage of two European spacecraft in orbit around Mars, the European Space Agency is currently preparing an experiment that consists of mutual radio occultations between Mars Express and the ExoMars Trace Gas Orbiter. Both spacecraft use UHF transceivers that are included primarily for communication between landers on the surface of Mars and the spacecraft, where the spacecraft act as relay orbiters to pass the data from the landers on to Earth. Therefore, these mutual occultations will be performed in the UHF range (centered around a frequency of 400 MHz). The feasibility of this technique on UHF was demonstrated between the NASA Mars Odyssey and Mars Reconnaissance Orbiters [Ao et al., 2015]. In this presentation, the advantages and challenges of this technique over the traditional spacecraft to Earth occultation measurements, the plans for conducting these experiments with Mars Express and the Trace Gas Orbiter, and the envisaged data processing technique will be briefly reviewed. Before the data becomes available, and in order to prepare the data processing, a simulation-based strategy has been adopted to implement an algorithm able to retrieve vertical electron density profiles from Doppler shift measurements. More specifically, as a first step, simulated spacecraft orbits are calculated and a Chapman function is used to obtain the electron density of the Martian ionosphere. Subsequently, a numerical 3D ray-tracing algorithm [Kashcheyev et al., 2012] is applied to compute ray trajectories in the presence of the ionosphere and the relevant Doppler shift time series corresponding to the simulated radio occultation event. Then, assuming a spherical symmetry [Fjeldbo et al., 1971] for the ionosphere electron density, the (excess) Doppler data are converted to bending angles and impact parameters. Finally, the bending angle profile is inverted (through Abel integral) to a vertical refractivity profile, which, in turn, provides information about the ionospheric electron density. For completeness, the simulation described above has been carried out with an exponential refractivity function defining the neutral atmosphere alone and with both the Chapman and the exponential refractivity functions to simulate the whole atmosphere of Mars. The first results obtained by means of the mentioned approaches will be presented, with particular focus on the retrieval of the ionospheric electron density profiles. References Ao, C. O., C. D. Edwards Jr., D. S. Kahan, X. Pi, S. W. Asmar, and A. J. Mannucci (2015), A first demonstration of Mars crosslink occultation measurements, Radio Sci., 50, 997–1007, doi:10.1002/2015RS005750. Fjeldbo, G., A. J. Kliore, and V. R. Eshleman (1971), The neutral atmosphere of Venus as studied with the Mariner V radio occultation experiments, Astron. J., 76, 123–140. Kashcheyev, A., B. Nava, and S. M. Radicella (2012), Estimation of higher-order ionospheric errors in GNSS positioning using a realistic 3-D electron density model, Radio Sci., 47, RS4008, doi:10.1029/2011RS004976

Spacecraft-to-spacecraft radio occultations experiments are being conducted at Mars between Mars Express (MEX) and Trace Gas Orbiter (TGO), the first ever extensive inter-spacecraft occultations at a planet other than Earth. Here we present results from the first 83 such occultations, conducted between 2 Nov 2020 and 5th of July 2023. Of these, 44 observations have to-date resulted in the extraction of vertical electron density profiles. These observations are the successful results of a major feasibility study conducted by the European Space Agency to use pre-existing relay communication equipment for radio science purposes. Mutual radio occultations have numerous advantages over traditional spacecraft-to-ground station occultations. In this work, we demonstrate how raw data are transformed into electron density values and validated with models and other instruments.

Theoretical modelling of the local ionospheric medium (LIM) is made difficult by the occurrence of irregular ionospheric behaviours at many space and time scales, making prior hypotheses uncertain. Investigating the LIM from scratch with the tools of dynamical system theory may be an option, using the vertical total electron content (vTEC) as an appropriate tracer of the system variability. An embedding procedure is applied to vTEC time series to obtain the finite dimension (m∈N) of the phase space of an LIM-equivalent dynamical system, as well as its correlation dimension (D2) and Kolmogorov entropy rate (K2). In this paper, the dynamical features (m,D2,K2) are studied for the vTEC on the top of three GNSS stations depending on the time scale (τ) at which the vTEC is observed. First, the vTEC undergoes empirical mode decomposition; then (m,D2,K2) are calculated as functions of τ. This captures the multi-scale structure of the Earth’s ionospheric dynamics, demonstrating a net distinction between the behaviour at τ≤24h and τ≥24h. In particular, sub-diurnal-scale modes are assimilated to much more chaotic systems than over-diurnal-scale modes.

The Brazilian equatorial and low-latitude regions are subject to various dynamic and electrodynamics processes. As a result, iono-spheric modeling over this region remains a challenge. In this article, we present the results of first observation of data ingestion into the climatological model, NeQuick during both quiet and disturbed conditions over Brazil. The variation of the daily F10.7 solar radio flux, the main driver of the NeQuick model, strongly influences its performance in both space and time, especially during high solar activ-ity. With data ingestion, using the local level of ionization, NeQuick's performance can be improved. We developed an algorithm to obtain the local effective ionization parameters (Az1 and Az2) using a single station, in the equatorial trough and low-latitude regions, which are subsequently used in the NeQuick to reproduce vTEC at co-located stations. The model's input (effective ionization level) was obtained when the modeled vTEC best fits the measured vTEC at the reference stations Maraba (5.35 degrees S, 49.11 degrees S, dip lat.: 3.06 degrees S; MABA) and Ourinhos (22.93 degrees S, 49.88 degrees S, dip lat.: 17.42 degrees S; OURI). Statistical results show that the model's performance greatly improves after data ingestion, reproducing vTEC at all latitudes close to the reference stations in 2014. We found that NeQuick improved by 71 %, 74 %, 83 %, and 69 % after ingestion during the storm periods of 17-21 February, 10-14 April, 6-10 June, and 23-27, December in the low-latitude region at SJSP. Using the Az1 values obtained at MABA and Az3 at SALU during July 2014, NeQuick reproduces the critical frequency of the F2 layer with a percentage improvement of approximately 20 % and 37 % respectively.(c) 2023 COSPAR. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/ by-nc-nd/4.0/).

Modelling the Earth’s ionosphere is a big challenge, due to the complexity of the system. Different first principle models have been developed over the last 50 years, based on ionospheric physics and chemistry, mostly controlled by Space Weather conditions. However, it is not understood in depth if the residual or mismodelled component of the ionosphere’s behaviour is predictable in principle as a simple dynamical system, or is conversely so chaotic to be practically stochastic. Working on an ionospheric quantity very popular in aeronomy, we here suggest data analysis techniques to deal with the question of how chaotic and how predictable the local ionosphere’s behaviour is. In particular, we calculate the correlation dimension D2 and the Kolmogorov entropy rate K2 for two one-year long time series of data of vertical total electron content (vTEC), collected on the top of the mid-latitude GNSS station of Matera (Italy), one for the year of Solar Maximum 2001 and one for the year of Solar Minimum 2008. The quantity D2 is a proxy of the degree of chaos and dynamical complexity. K2 measures the speed of destruction of the time-shifted self-mutual information of the signal, so that K2−1 is a sort of maximum time horizon for predictability. The analysis of the D2 and K2 for the vTEC time series allows to give a measure of chaos and predictability of the Earth’s ionosphere, expected to limit any claim of prediction capacity of any model. The results reported here are preliminary, and must be intended only to demonstrate how the application of the analysis of these quantities to the ionospheric variability is feasible, and with a reasonable output.

Studies on the irregularities of the ionosphere during disturbed geomagnetic conditions are fundamental to understanding the complex dynamics taking place in the upper atmosphere. In this work, different data sources are used to study the ionosphere effects of two moderate geomagnetic storms, 26–27 February 2014 and 17–18 September 2021, over the Iberian Peninsula. Data are obtained from digital ionosondes in Spain, Italy and Greece; the Global Navigation Satellite System (GNSS) derived Total Electron Content (TEC) and Rate Of TEC Index (ROTI) from several receiver stations in Spain, Portugal and Morocco; and the UPC Quarter-of-an-hour time resolution Rapid GIM (UQRG), vertical TEC global ionosphere maps (GIMs), produced at 15 min intervals by the Universitat Politecnica de Catalunya (UPC, Spain). This analysis showed that, during the two moderate storms, spread-F and high values of ROTI, indicating the presence of irregularities, are found in a very localized area (Southern Iberian Peninsula and northwest Africa) and local times (night-time). However, no irregularities are found eastwards and northwards of the location indicated. We propose some possible explanations for these observations for both the storms, one of them related to the position of the Equatorial Ionosphere Anomaly (EIA) and the other one attributed to the Perkins’ instabilities.