Wiley

Space Weather

Published by Wiley and American Geophysical Union

Online ISSN: 1542-7390

Disciplines: Earth and space science

Journal websiteAuthor guidelines

Top-read articles

118 reads in the past 30 days

GNSS receiver system assembled and TOPGNSS antenna.
EclipseNB station locations and the 8 April 2024 solar eclipse totality path calculated at a height of 300 km.
Uncalibrated STEC values obtained for the GPS constellation using the low‐cost and scientific grade receivers during DOY = 344 (panel a), and DOY = 335 (panel c). Panels (b) and (d) show the statistics of the STEC difference.
The number of cycle slips experienced by the low‐cost (orange) and the scientific grade (blue) receivers during DOY = 344 (panel a) and DOY = 335 (panel b).
VTEC values obtained using the low‐cost and scientific grade receivers during DOY = 344 (panel a), and DOY = 335 (panel c). Panels (b) and (d) show the statistics on the VTEC difference.

+3

EclipseNB: A Network of Low‐Cost GNSS Receivers to Study the Ionosphere

January 2025

·

118 Reads

·

·

C. Watson

·

[...]

·

R. B. Langley
Download

Aims and scope


Space Weather is an open access journal that publishes original research articles and commentaries devoted to understanding and forecasting space weather and other interactions of solar processes with the Earth environment, and their impacts on telecommunications, electric power, satellite navigation, and other systems.

Recent articles


On the Geoelectric Field Response to the SSC of the May 2024 Super Storm Over Europe
  • Article
  • Full-text available

February 2025

·

8 Reads

·

D. M. Oliveira

·

G. D’Angelo

·

[...]

·

E. Zesta

Solar variability can lead to significant disturbances, such as coronal mass ejections (CMEs). A CME impacting the Earth's magnetosphere often causes geomagnetic storms that affect not only the magnetosphere but also the ionosphere, the upper atmosphere, and even the ground. During extreme events, rapidly changing geomagnetic fields can create strong geomagnetically induced currents (GICs) at ground. These GICs can severely impact human technology, causing damage to high‐voltage power transformers and leading to power outages, as well as corrosion in oil and gas pipelines. On 10 May 2024, the most intense geomagnetic storm since the Halloween 2003 storm impacted Earth's environment, causing auroras to appear at much lower latitudes than usual in both the northern and southern hemispheres. This study investigates the effects of geomagnetically induced electric fields (GIEs), and hence GICs, during the sudden storm commencement (SSC) of the geomagnetic storm on 10 May 2024, over Europe, using the European quasi‐Meridional Magnetometer Array ground magnetometers. Despite the magnetometer array being placed in the late afternoon (18:00 LT), the combined influence of a strong solar wind dynamic pressure amplitude (P∼22nPa) (P22nPa)(P\sim 22nPa) and a significant, long‐lasting southward interplanetary magnetic field (IMF) (Bz,IMF∼−25 Bz,IMF25{B}_{z,IMF}\sim -25nT) resulted in strong SSC amplitudes (∼180 180{\sim} 180nT) at mid‐low latitudes (λ∼57° λ57\lambda \sim 57{}^{\circ}N). Results suggest that the CME‐driven shock inclination in the meridional plane leads to high GIE driven only at high latitudes. In addition, the decomposition of the SSC disturbance field at ground into ionospheric (DP field) and magnetospheric (DL field) origin contribution should commend input to GIEs (and hence to GICs) from both DL and DP fields, rather than ionospheric current alone.


Temporal variations in (a) solar wind speed, (b) interplanetary magnetic field (IMF) Bz and By components, (c) AE index, (d) SYM‐H index, (e) equatorial ionospheric eastward electric field at longitude 225°E obtained from model calculation, ROTI index in (f) MKEA and (g) THTI, and (h) Kp index during the period of 20 and 21 November 2003. The yellow shaded areas represent the period of interest for this study.
Geographic latitudinal profiles of (a) the ground tracks during 19 (red) and 21 (blue) November 2003, ionospheric ion density (b) during 19 November 2003 and (c) during 21 November 2003, (d) vertical ion drift velocity (Vz), and (e) O⁺ proportion. Vertical shaded areas with the same color indicate the corresponding geomagnetic conjugate plasma depletion. The observations during 19 November 2003 are displayed as quiet time reference. The geographic latitude (Glat), geomagnetic latitude (Mlat, red for 19 November 2003 and blue for 21 November 2003), and local time (LT) are annotated at the bottom of the figure. Note that the starting‐ending orbit times of the DMSP F13 satellite in Universal Time (UT) are also indicated in (b) and (c).
Overview of ionospheric disturbances in the East Pacific during the recovery phase of super storms on 21 November 2003. (a) The regional map at the 190–260°E longitude sector, including the trajectories of DMSP F13, ROCSAT‐1, CHAMP, and GRACE satellites, and the locations of ground‐based GPS station at MKEA (red star) and THTI (blue circle). ROCSAT‐1 observations of ionospheric ion density are displayed along the Rocsat‐1 orbits with the starting‐ending universal time and Roman numerals. The colored lines on DMSP F13 trajectories represent plasma depletions in both hemispheres. The dark red areas indicate plasma depletions in the Southern Hemisphere, projected along geomagnetic field lines to the Northern Hemisphere, highlighting strong geomagnetic conjugacy. Geographic latitudinal profiles of the ionospheric ion density as measured by (b) CHAMP, two GRACE orbits (c) GRACE 2, (d) GRACE 1, and (e) DMSP F13, using the corresponding colors in panel (a). Graphs contain information about the starting‐ending universal time (UT) and the direction of satellite trajectories marked by black arrows.
TIE‐GCM simulations of vertical plasma drifts (colors) and thermospheric winds (black arrows) at an altitude of 500 km with an interval of 1 hr from 8 UT to 15 UT on 21 November 2003, as provided by CCMC. The geographical latitude ranges from 50°S to 50°N. Red and blue solid lines indicate the sunrise and sunset terminators, respectively.
Observations of Unusual Postsunrise Interhemispheric Geomagnetic Conjugate Super Plasma Depletions at Midlatitudes During the Recovery Phase of the November 2003 Superstorm

February 2025

·

3 Reads

Hailun Ye

·

Xianghui Xue

·

Yang‐Yi Sun

·

[...]

·

Xiankang Dou

Plain Language Summary Ionospheric F region plasma irregularities, known as plasma depletions or bubbles, frequently occur at equatorial and low latitudes after sunset and can sometimes extend to mid‐latitudes under certain conditions. Typically, during the main phase of a magnetic storm, the prompt penetration electric field (PPEF) combines with the normal sunset pre‐reversal enhancement (PRE) to create favorable conditions for the formation of strong equatorial plasma bubbles (EPBs), which can extend along the geomagnetic field lines to midlatitudes. However, our study reports an extreme ionospheric plasma depletion event over the Eastern Pacific from postmidnight to early morning during the superstorm recovery phase on 21 November 2003, occurring more than 12 hr after the end of the main phase. In this case, the observed geomagnetic conjugate plasma depletion may reach an apex altitude of more than 6,000 km over the geomagnetic equator, making it the largest super plasma depletion of equatorial origin observed after sunrise. Interestingly, these super plasma depletions appeared to be sustained and growing due to upward drift in the Northern Hemisphere during the early morning of 21 November. This suggests that storm‐induced postmidnight disturbance dynamo electric field (DDEF) can trigger a similar intensive latitudinal extension of midlatitude plasma depletions as the PPEF effect does after sunset.


A Novel Short‐Term Prediction Model for Regional Equatorial Plasma Bubble Irregularities in East and Southeast Asia

February 2025

·

24 Reads

Equatorial plasma bubble (EPB) irregularities can significantly impact satellite‐based communication and navigation systems. Accurate prediction of EPB occurrence is essential for mitigating these impacts. Using the GNSS receiver network and ionosonde data from East and Southeast Asia during 2010–2021, and the rate of TEC change index to characterize the occurrence of EPB irregularities, we developed a novel Spatio‐Temporal deep learning model for regional EPB irregularities short‐term Prediction (STEP). The model integrates the convolutional neural network and long short‐term memory (LSTM) network, together with attention mechanisms, to capture both spatial and temporal features of regional ionospheric irregularities. The results show that for 5‐min forecast, the STEP model achieves a root mean square error (RMSE) of 0.062 TECU/min and an R² of 0.818, reducing RMSE by 19.48% compared to LSTM and 27.06% compared to gated recurrent unit model. For 60‐min prediction, the STEP model can still achieve reasonable accuracy with an RMSE of 0.110 TECU/min and an R² of 0.482, showing significant improvement over traditional models. The equatorial F layer height and regional TEC fluctuations were identified as the most critical factors for predicting the generation and duration of EPB irregularities, respectively. The spatial and temporal distributions of EPB irregularities, including their latitudinal variation and delayed onset after sunset, and the occurrence across different days in East and Southeast Asia, were well predicted by the STEP. It is expected that the STEP model would provide a valuable tool for improving the resilience of GNSS against ionospheric scintillations induced by EPB irregularities.


Solar Wind and Magnetospheric Conditions for Satellite Anomalies Attributed to Shallow Internal Charging

February 2025

·

6 Reads

Satellite charging in Earth's magnetospheric plasma and radiation belts frequently causes operational anomalies. A recent study of a frequent Space Wire anomaly on GOES linked it to shallow internal charging by 100–300 keV electrons. Solar wind and magnetospheric conditions during a period spanning solar minimum (2017–2021) are analyzed to gain further insight into the anomalies. The anomalies are divided into two groups, with inter‐anomaly intervals shorter than 2 days (clustered) and longer (background). The clusters sometimes exhibit a clear 27‐day recurrence. The magnetic local time (MLT) distributions for background anomalies are statistically uniform, unlike those for clustered anomalies, which peak prior to local noon. The clustered anomaly distributions with respect to Kp are similar to those for surface charging, indicating enhanced plasma sheet access to geostationary orbit. The maximum 100–300 keV fluxes during injections are similar to published extreme fluxes. Through superposed epoch analysis and comparison with published lists of high‐speed streams, clustered anomalies are determined to occur during high‐speed streams with elevated solar wind speed, lower number density, and weakly negative interplanetary magnetic field Bz Bz{B}_{z}. The MLT and Kp dependencies of the clustered anomalies may indicate charge accumulating in a thin dielectric with a time constant less than 1 day. The background anomalies, occurring uniformly in local time and varying slowly between solar rotation periods, may indicate a distinct charging location with a time constant greater than a solar rotation.


The Need for a Sub‐L1 Space Weather Research Mission: Current Knowledge Gaps on Coronal Mass Ejections

February 2025

·

7 Reads

Over the past decades, missions at the L1 point have been providing solar wind and interplanetary magnetic field measurements that are necessary for forecasting space weather at Earth with high accuracy and a lead time of a few tens of minutes. Improving the lead time, while maintaining a relatively high level of accuracy, can be achieved with missions sunward of L1, so‐called sub‐L1 monitors. However, too much is unknown to plan for sub‐L1 monitors as operational missions: both the orbital requirements of such missions, and the achievable accuracy of forecasts based on their measurements have not been quantitatively defined. We review here some proposed mission concepts and explain the knowledge gaps related to coronal mass ejections (CMEs) that require a space weather research or science mission. We first show how STEREO‐A measurements in 2023 can be used as a proof of concept of the use of sub‐L1 monitor slightly off the Sun‐Earth line to forecast the Dst index. We then highlight that separations of ≲10° 10\lesssim 10{}^{\circ} are needed to ensure that CMEs measured by a sub‐L1 monitor impact Earth. Next, we show that measurements with angular separations of ≲0.35° 0.35\lesssim 0.35{}^{\circ} have negligible errors but separations of a few degrees can result in significant errors in lead time and in the forecasted magnetic field strength of CMEs. We also discuss how CME evolution over the last 0.05–0.2 au before impacting Earth is strongly under‐constrained and needs to be better understood before using measurements of sub‐L1 monitors for real‐time space weather forecasting.


Illustration of the half‐wave rectified vBs $v{B}_{s}$ coupling function used as input for the WINDMI model throughout a day. (a) Solar wind velocity along the Sun‐Earth line. (b) IMF Bz ${B}_{z}$ indicating the North‐South magnetic field (in GSM coordinate system), with the blue dashed line denoting the 0 value. (c) Rectified input obtained by multiplying solar wind velocity and southward‐directed magnetic field, supplemented with a 4 kV threshold voltage.
Figure: Illustration of the WINDMI model's trigger function θI−Ic $\theta \left(I-{I}_{c}\right)$ during a day. (a) Magnetotail current I $I$ and critical current Ic ${I}_{c}$ (blue dashed line). (b) Trigger function θ $\theta $, increasing when I $I$ crosses Ic ${I}_{c}$ and decreasing when below. (c) Changes in Region 1 field‐aligned current values, with green shading indicating periods of high θ $\theta $ values.
Example of merging substorm onset times obtained from different substorm lists from the SuperMAG website. The figure displays a 1.5‐hr period. Both panels show the SML index in nT. Panel (a) features three dotted yellow vertical lines within approximately 5 min of each other, corresponding to Forsyth, Ohtani, and Newell (from left to right). Panel (b) illustrates the merged substorm onset as one dotted vertical line at the average of the three times shown in Panel (a).
WINDMI output and SML index on 17 March 1998. (a) Solar wind input vBs $v{B}_{s}$ (black line) and IMF Bz ${B}_{z}$ (blue shades) with 0 nT reference line. (b) Magnetotail current I $I$ with critical threshold Ic ${I}_{c}$ (blue dashed line). (c) R1 current I1 ${I}_{1}$ and trigger function θ $\theta $ in green. (d) SML indices with substorm onsets as dashed lines, color‐coded by the number of concurring methods, and labeled by detecting authors' abbreviations.
WINDMI output and SML index on 7 January 2000. (a) Solar wind input vBs $v{B}_{s}$ (black line) and IMF Bz ${B}_{z}$ (blue shades) with 0 nT reference line. (b) Magnetotail current I $I$ with critical threshold Ic ${I}_{c}$ (blue dashed line). (c) R1 current I1 ${I}_{1}$ and trigger function θ $\theta $ in green. (d) SML indices with substorm onsets as dashed lines, color‐coded by the number of concurring methods, and labeled by detecting authors' abbreviations.
Substorm Identification With the WINDMI Magnetosphere ‐ Ionosphere Nonlinear Physics Model

February 2025

·

11 Reads

We investigate the applicability and performance of the plasma physics based WINDMI model to the analysis and identification of substorm onsets. There are several substorm onset criteria that have been developed into event lists, either from auroral observations or from auroral electrojet features. Five of these substorm onset lists are available at the SuperMAG website. We analyze these lists, aggregate them and use the WINDMI model to assess the identified events, emphasizing the loading/unloading mechanism in substorm dynamics. The WINDMI model employs eight differential equations utilizing solar wind data measured at L1 by the ACE satellite as input to generate outputs such as the magnetotail current, the ring current and the field‐aligned currents (FACs). In particular, the WINDMI model current output I1 I1{I}_{1} represents the westward auroral electrojet, which is related to the substorm SML index. We analyze a decade of solar wind and substorm onset data from 1998 to 2007, encompassing 39,863 onsets. Our findings reveal a significant correlation, with WINDMI‐derived enhancements in FAC coinciding with the identified substorm events approximately 32% of the time. This suggests that a substantial proportion of substorms may be attributed to solar wind driving that results in the loading and unloading of energy in the magnetotail.


Assessing 1‐Second ROTI for Ionospheric Perturbation Monitoring Using Real‐Time Multi‐GNSS Data in China

February 2025

·

68 Reads

With the onset of solar activities of the 25th solar cycle, ionospheric perturbations have become increasingly frequent, posing significant challenges to various applications of Global Navigation Satellite Systems (GNSS). The Rate of Total Electron Content (TEC) Index (Rate of TEC Index (ROTI)), commonly derived from GNSS data sampled at 30‐s intervals, fails to detect ionospheric irregularities smaller than the first Fresnel scale due to its low temporal resolution. This study investigates the efficacy of using 1‐s sampling intervals of multi‐GNSS (GPS, GLONASS, Galileo and BDS) data for monitoring ionospheric perturbations, comparing it to the conventional 30‐s sampling approach. Using real‐time multi‐GNSS observations in China, we computed ROTI values at both 1‐s (1s‐ROTI) and 30‐s (30s‐ROTI) intervals. Results indicate that 1s‐ROTI demonstrates higher magnitude and larger inconsistency across different GNSS constellations compared to 30s‐ROTI, with peak 1s‐ROTI values reaching 10 TECU/min and a maximum discrepancy of 1.0 TECU/min between different GNSS constellations. The inconsistency in 1s‐ROTI was found to correlate with different receivers types (e.g., Septentrio, Unicore, Trimble and Leica). Additionally, 1s‐ROTI exhibits better consistency and higher correlation with scintillation indices from Ionospheric Scintillation Monitoring Receivers (ISMR), with overall correlation coefficient exceeding 0.85 for all systems except for GLONASS. To mitigate inconsistencies in 1s‐ROTI, we propose different GNSS‐ROTI indices, with ROTImean and ROTIele showing the best performance in terms of correlation with scintillation indices and reflecting GNSS positioning performance degradation. The potential of 1s‐ROTI is highlighted for improving small‐scale ionospheric irregularities monitoring and indicating the GNSS positioning performance, especially in regions lacking ISMR.


RA‐ConvLSTM: Recurrent‐Architecture Attentional ConvLSTM Networks for Prediction of Global Total Electron Content

February 2025

·

15 Reads

The ionosphere poses a significant source of error in satellite‐based navigation systems for aviation and radio communication applications. Accurate estimation of the total electron content (TEC) can effectively mitigate the impact of such errors. However, constrained by observational techniques, the acquisition of global ionospheric TEC in practical applications relies heavily on high‐precision forecasting products. In this study, we construct a global ionospheric forecasting model based on the global ionospheric TEC products disseminated by the International GNSS Service (IGS) and deep learning. We incorporate an attention module to extract global spatiotemporal features from historical ionospheric data and employ these features to predict the TEC values over the next 24 hr. Additionally, we select a long short‐term memory (LSTM) model and a ConvLSTM model as baseline models for comparison, conducting experiments under varying solar activity conditions. The experimental results demonstrate that RA‐ConvLSTM model outperforms the other two models in quantifying the performance of the models. During high solar activity years, the bias and Root Mean Square Error (RMSE) of RA‐ConvLSTM model are −0.0298 TECU and 3.8980 TECU, respectively, while during low solar activity years, these values are 0.0905 TECU and 1.5059 TECU, marking a notable improvement over the comparative models. Furthermore, by contrasting the precision of the three forecasting models during geomagnetic storms, the RA‐ConvLSTM model exhibits the least fluctuations in accuracy, indicative of a higher degree of stability in its forecasting outcomes.


Multi‐Scale Intense Geoelectric and Geomagnetic Field Perturbations Observed After an Interplanetary Magnetic Field Turning

February 2025

·

30 Reads

Intense geoelectric fields during geomagnetic storms generate geomagnetically induced currents in power grids and other infrastructure, necessitating an understanding of their causes, for example, through coordinated space and ground observations. This study investigates localized intense geoelectric (E) and geomagnetic (B) field perturbations following an Interplanetary Magnetic Field (IMF) turning during a geomagnetic storm on 25 October 2011. Observations from EarthScope magnetotelluric sites in the upper Midwest United States revealed shorter period (∼ {\sim} 1 min) ultra‐low‐frequency (ULF) waves superimposed on longer period (∼ {\sim} 10 min) perturbations in both E and B fields. These sites, located at ∼ {\sim} 19 hr magnetic local time and 56−57° 565756-57{}^{\circ} magnetic latitude, recorded large amplitude E and B perturbations. Ground‐based all‐sky imagers showed auroral brightening with sunward and poleward propagation, while upstream spacecraft linked the perturbations to an IMF turning and solar wind dynamic pressure impulse. The longer‐period E and B field perturbations likely stem from localized ionospheric currents tied to substorm auroral activity post‐IMF turning. The combination of ionospheric currents, ULF waves, and the Earth's varying conductivity produces intense geoelectric fields of ≥ {\ge} 2 V/km in the upper Midwest. A comparison using input data and software compatible with the NOAA/USGS geoelectric field nowcast model revealed its limitations in capturing such events due to the temporal and spatial resolution of the underlying data. Using 1‐s geomagnetic field data can improve geoelectric field models by capturing short‐period and large spatial scale waves, although localized magnetic perturbations remain underestimated due to insufficient ground magnetometer density.



Cross Correlation Between Plasmaspheric Hiss Waves and Enhanced Radiation Levels at Aviation Altitudes

February 2025

·

16 Reads

Enhanced radiation in the Earth's atmosphere can pose serious hazards to pilots, aircraft passengers, and commercial space travelers. Recent results have shown, statistically, that there is a strong correlation between dose rates observed by Automated Radiation Measurements for Aerospace Safety (ARMAS) instruments at aviation altitudes (>9 km) and plasmaspheric hiss wave power measured by NASA's Van Allen Probes within the inner magnetosphere. Plasmaspheric hiss waves play a very important role in removing energetic electrons from the Earth's radiation belts by precipitating them into the upper atmosphere. These relativistic electrons generally drift eastwards along closed magnetic drift shells. In this study, we use magnetic conjunction events between ARMAS and the Van Allen Probes to analyze the causality between plasmaspheric hiss waves and enhanced radiation observed at aviation altitude. We specifically study how the size of the conjunction window and a shift in L and MLT of the conjunction window affect the correlation between dose rates and plasmaspheric hiss wave power. This is to determine if the observed enhanced radiation at aviation altitude is indeed caused by the plasmaspheric hiss waves in the inner magnetosphere. The results show that the enhanced radiation levels are only correlated with plasmaspheric hiss waves within conjunction windows of −1 ≤ {\le} L ≤ {\le} 1 and 0 ≤ {\le} MLT ≤ {\le} 2. The correlation between dose rate and hiss wave power increases slightly if ARMAS is shifted approximately 1 hr in MLT to the east of the Van Allen Probes, consistent with the drift trajectory of the electrons precipitating into the atmosphere.


Impact of Ionospheric Scintillations on GNSS Availability and Precise Positioning

February 2025

·

38 Reads

For many applications where accuracy, availability, and integrity are essential, such as geodetic positioning and civil aviation, Global Navigation Satellite Systems (GNSS) are indispensable. However, GNSS signal dependability is severely hampered by ionospheric disturbances, especially equatorial plasma bubbles (EPBs), particularly in equatorial latitudes. To better understand how ionospheric irregularities impact GNSS signals across several constellations (GPS, GLONASS, Galileo, BeiDou, and Satellite‐Based Augmentation Systems) and frequencies, this study examines ionospheric amplitude scintillation. The objective is to comprehend how these anomalies simultaneously affect GNSS performance in high scintillation activity conditions. The research focuses on the effects of amplitude scintillation during periods of high solar activity by analyzing data from four stations across Brazil. The analysis identified the most critical hours for scintillation events, between 20:00 and 23:59 LST, where up to 13 satellites were simultaneously affected at PRU2, resulting in a notable drop in positioning accuracy. This was further reflected in the degradation of Position Dilution of Precision values, which exceeded 5 in approximately 38% of the cases at Presidente Prudente and São José dos Campos, indicating reduced confidence in positioning accuracy during severe scintillation events. Additionally, the study confirms that lower‐frequency signals (L2, G2, B2) are more susceptible to scintillation than higher‐frequency signals (L1, G1, E1). Despite multi‐constellation capabilities, the simultaneous impact of EPBs on multiple GNSS signals leads to degraded satellite availability and positioning accuracy, especially in regions with high electron density. These findings highlight the need for improved mitigation strategies in multi‐constellation systems to enhance GNSS reliability in equatorial regions.


Decent Estimate of CME Arrival Time From a Data‐Assimilated Ensemble in the Alfvén Wave Solar Atmosphere Model (DECADE‐AWSoM)

January 2025

·

24 Reads

Forecasting the arrival time of Earth‐directed coronal mass ejections (CMEs) via physics‐based simulations is an essential but challenging task in space weather research due to the complexity of the underlying physics and limited remote and in situ observations of these events. Data assimilation techniques can assist in constraining free model parameters and reduce the uncertainty in subsequent model predictions. In this study, we show that CME simulations conducted with the Space Weather Modeling Framework (SWMF) can be assimilated with SOHO LASCO white‐light (WL) observations and solar wind observations at L1 prior to the CME eruption to improve the prediction of CME arrival time. The L1 observations are used to constrain the model of the solar wind background into which the CME is launched. Average speed of CME shock front over propagation angles are extracted from both synthetic WL images from the Alfvén Wave Solar atmosphere Model (AWSoM) and the WL observations. We observe a strong rank correlation between the average WL speed and CME arrival time, with the Spearman's rank correlation coefficients larger than 0.90 for three events occurring during different phases of the solar cycle. This enables us to develop a Bayesian framework to filter ensemble simulations using WL observations, which is found to reduce the mean absolute error of CME arrival time prediction from about 13.4 to 5.1 hr. The results show the potential of assimilating readily available L1 and WL observations within hours of the CME eruption to construct optimal ensembles of Sun‐to‐Earth CME simulations.


Ionospheric Response to the 24–27 February 2023 Solar Flare and Geomagnetic Storms Over the European Region Using a Machine Learning–Based Tomographic Technique

January 2025

·

19 Reads

Two solar flares accompanied by coronal mass ejections (CMEs) occurred on 24–25 February (DOY 055–056), 2023, resulting in a large magnetic storm on DOY 058. We reconstructed the ionospheric electron density (IED) in Europe to analyze the spatial distribution of ionosphere and its temporal evolution during this period. Computerized ionospheric tomography based on machine learning (CIT‐ML) was used to predict the IEDs of unobserved voxels. The IEDs were examined using observation data from the Swarm satellite. The CIT‐ML accuracy was 28.3% higher than the improved algebraic reconstruction technique with relaxation factor automatic search technology (IART‐AS), which effectively improved the typical ill‐posed problem of CIT. The first flare generated the Bz component of the interplanetary magnetic field (IMF), which continued southward for 13 hr, causing a small magnetic storm before the second flare occurred, resulting in an increased nighttime IED and nighttime medium‐scale traveling ionospheric disturbance (MSTID). The vertical total electron content (VTEC) and IED declined in the early stages of the main phase of the large magnetic storm, but later increased, indicating that negative‐positive biphasic storms were occurring in the ionosphere that altered the ionospheric daily cycle, resulting in the peak of the ionosphere being advanced by approximately 1.5 hr. The storms also caused nighttime MSTIDs during the main phase (DOY 057 at night) and the recovery phase (DOY 058 at night). To investigate the mechanisms of these results, we conducted a term analysis of the ion continuity equation using the Thermosphere Ionosphere Electrodynamics General Circulation Model (TIEGCM). The analysis showed that ambipolar diffusion driver nighttime MSTIDs during flares, while increased geomagnetic disturbances amplify the effects of neutral wind transport, E × B drift and chemical reactions during magnetic storms. These combined effects offset the alternating positive and negative structures induced by ambipolar diffusion, becoming the main cause of electron density variations during ionospheric storms.


EclipseNB: A Network of Low‐Cost GNSS Receivers to Study the Ionosphere

January 2025

·

118 Reads

Plain Language Summary The Earth's ionosphere affects the performance of communication and navigation systems, such as global navigation satellite systems (GNSS). Improving the knowledge of the state of the ionosphere is a key factor toward increasing the operational capabilities of such systems. One of the most widespread tools for obtaining the ionospheric information is GNSS itself. GNSS are designed to compensate for the ionospheric delay utilizing the dispersive nature of the ionosphere. Using multiple frequencies makes it possible to estimate the ionospheric delay. Retrieving information from dense networks of GNSS receivers allows for reconstruction of ionospheric electron density with high spatial and temporal resolution. One of the main barriers in using dense GNSS networks is the cost of multi‐frequency GNSS receivers. This work aims to demonstrate that dense networks of low‐cost dual‐frequency GNSS receivers can be used to retrieve ionospheric information with almost the same level of accuracy as scientific‐grade GNSS receivers. A network of 15 GNSS receivers was developed to study the structure and dynamic behavior of the ionosphere, including its response to the total solar eclipse in April 2024. Ionospheric observations during the solar eclipse and the extreme geomagnetic storm in May 2024 are presented.


Forecasting High‐Speed Solar Wind Streams From Solar Images

January 2025

·

37 Reads

The solar wind, a stream of charged particles originating from the Sun and transcending interplanetary space, poses risks to technology and astronauts. In this work, we develop a prediction model to forecast the solar wind speed (SWS) at the Earth. We focus on high‐speed streams (HSSs) and their solar source regions, coronal holes. As input, we use the coronal hole area, extracted from solar extreme ultraviolet (EUV) images and mapped on a fixed grid, as well as the SWS 27 days before. We use a polynomial regression model and a distribution transformation to predict the SWS with a lead time of 4 days. Our forecast achieves a root mean square error (RMSE) of 68.1 km/s for the SWS prediction and an RMSE of 76.8 km/s for the HSS peak velocity prediction for 2010 to 2019. We also demonstrate the applicability of our model to the current solar cycle 25 in an operational setting, resulting in an RMSE of 80.3 km/s and an HSS peak velocity RMSE of 92.2 km/s. The study shows that a small number of physical features explains most of the solar wind variation, and that focusing on these features with simple machine learning algorithms even outperforms current approaches based on deep neural networks and MHD simulations. In addition, we explain why the typically used loss function, the mean squared error, systematically underestimates the HSS peak velocities, aggravates operational space weather forecasts, and how a distribution transformation can resolve this issue.


Storm‐Time Ring Current Plasma Pressure Prediction Based on the Multi‐Output Convolutional Neural Network Model

January 2025

·

35 Reads

The terrestrial ring current consists of particles with energy from several keV to 100 s of keV, and its enhancement will result in magnetic field depression, known as geomagnetic storms. The ring current is mainly composed of H⁺, O⁺, He⁺, and electrons, and there has been a longstanding debate regarding their relative contributions. In this study, we employed a multi‐output convolutional neural network to predict the storm‐time ring current plasma pressures of these particles. Taking solar wind parameters, interplanetary magnetic field (IMF) data, and geomagnetic indices with a time history of 3 days as input parameters, the model shows good performances for electron plasma pressure, H⁺ plasma pressure, He⁺ plasma pressure, and O⁺ plasma pressure in both quiet‐time and storm‐time periods, with high correlation coefficients and small root mean square errors between the measured and the predicted values. Our model successfully captures ring current enhancement during the storm main phase and species‐dependent decay during the recovery phase. Moreover, the model's ability to predict plasma pressures of different species simultaneously facilitates a comparative analysis of their respective contributions to the ring current. The storm event where our model was applied demonstrates that during the storm, the contribution of H⁺ decreases but still dominates, while the contribution of O⁺ increases dramatically, and electrons and He⁺ also play roles in some localized regions. The application of this model is in line with previous observations and simulations, which can be utilized for quantitative analysis of storm‐time ring current dynamics.


Radiation Impact of the Halloween GLE Events During the October–November 2003 Period

January 2025

·

42 Reads

Plain Language Summary Solar energetic particles can be created during strong solar eruptions. When these particles possess enough energy, they penetrate the Earth's atmosphere creating cascades of secondary particles which can be detected at the ground. We call these events ground‐level enhancements. Such events can have numerous space weather impacts that are hazardous to humans, such as damaging electronics and enhancing the radiation environment at high altitudes. During late October to early November 2003 several large solar eruptions were detected leading to one of the strongest geomagnetic storms recorded and three ground‐level enhancement events, colloquially known as the Halloween events. It is important to analyze the space weather effects, in this case, the radiation impact, of the Halloween events that occurred during these unique storm conditions to better prepare for similar future events. Using a new radiation tool and the recently derived spectra for the solar energetic particles during the Halloween events we determined the radiation impact at aviation altitudes. The radiation dose was found to have more than doubled above the typical levels during the Halloween events, however, the risk to human health remained relatively minor. Good agreement between the radiation model and flight measurements during the Halloween events was also found.


Comparative Hypothesis Testing of Auroral Ionospheric Layer Causing Global Navigation Satellite System Scintillation

January 2025

·

7 Reads

As Global Navigation Satellite System electromagnetic waves pass through the ionosphere, especially in auroral zones, ionospheric irregularities cause the waves to scintillate. Identification of the ionosphere scattering layer is an important factor in understanding the cause of scintillation. This work implements two techniques to determine whether signal scattering for Global Positioning System L1 and L2C signals might be in the E‐ or F‐layer. The first technique used is an updated process of Sreenivash et al. (2020, https://doi.org/10.1029/2018RS006779), in which the Poker Flat Incoherent Scatter Radar (PFISR) maximum electron densities and their uncertainties hypothesize the layer in which scattering has occurred. The density‐based method predicts a majority of F‐region scintillation events for 2014, with a majority of E‐region events found for 2015 to 2019. The second technique consists of using the ratio of the 630 (red) to the 428 nm (blue) intensity in optical all‐sky images (ASIs) to hypothesize the scattering layer with ASI. The decision threshold is set to 1.35 based on the GLobal airglOW model. From 2014 to 2018 174 events have both PFISR data and ASIs with clear viewing conditions and alignment to within 25° of magnetic zenith. There is an agreement between the two methods for 128 (74%) events.


Global Ionospheric Scintillation Estimation Based on Phase Screen Modeling From One‐Dimensional Satellite Data

January 2025

·

45 Reads

Ionospheric scintillations, which usually manifest as sudden, rapid fluctuations in radio wave signal phase and amplitude, challenge the reliability of satellite communication and navigation. Based on the single phase screen assumption, this study uses the one‐dimensional (1D) in‐situ plasma density data of ESA's Swarm constellation data to develop a three‐dimensional (3D) power spectrum of electron density perturbation and construct a model to estimate scintillations caused by small‐scale ionospheric plasma density irregularities. By deriving the turbulence strength (Cs Cs{C}_{s}) and calculating the amplitude scintillation index S4, the global distribution of ionospheric scintillation is derived. Scintillation from our model shows typical seasonal variations, with peaks during equinoxes at both high and low magnetic latitudes. For local time (LT) dependence, the scintillation at low magnetic latitudes peaks around 21:00 LT, while at high magnetic latitudes, the maximum occurrence appears around noon, with an asymmetry between the northern and southern hemispheres. In addition, positive correlations between scintillation occurrence and solar activity, as well as geomagnetic storms are observed, with higher magnetic latitudes more being affected by geomagnetic disturbances. These features of our model‐estimated scintillations agree well with the occurrence of small‐scale plasma density irregularities at different magnetic latitudes as reported by previous studies. Our study introduces a way to estimate the global coverage of ionospheric scintillation from in‐situ satellite measurements, which cannot be achieved by the ground‐based GNSS networks due to the lack of coverage in the ocean regions.


Understanding and Modeling the Dynamics of Storm‐Time Atmospheric Neutral Density Using Random Forests

January 2025

·

16 Reads

Atmospheric neutral density is a crucial component to accurately predict and track the motion of satellites. During periods of elevated solar and geomagnetic activity atmospheric neutral density becomes highly variable and dynamic. This variability and enhanced dynamics make it difficult to accurately model neutral density leading to increased errors which propagate from neutral density models through to orbit propagation models. In this paper we investigate the dynamics of neutral density during geomagnetic storms. We use a combination of solar and geomagnetic variables to develop three Random Forest machine learning models of neutral density. These models are based on (a) slow solar indices, (b) high cadence solar irradiance, and (c) combined high‐cadence solar irradiance and geomagnetic indices. Each model is validated using an out‐of‐sample data set using analysis of residuals and typical metrics. During quiet‐times, all three models perform well; however, during geomagnetic storms, the combined high cadence solar iradiance/geomagnetic model performs significantly better than the models based solely on solar activity. The combined model capturing an additional 10% in the variability of density and having an error up to six times smaller during geomagnetic storms then the solar models. Overall, this work demonstrates the importance of including geomagnetic activity in the modeling of atmospheric density and serves as a proof of concept for using machine learning algorithms to model, and in the future forecast atmospheric density for operational use.


Statistical Study on the Effect of Meridional Neutral Wind on the Occurrence of Post‐Sunset Equatorial Ionospheric Irregularities

January 2025

·

94 Reads

The navigation and radio communication systems experience significant disruptions due to post‐sunset equatorial ionospheric irregularities. There is ongoing debate regarding the impact of meridional/trans‐equatorial wind speed on these irregularities, though it is widely agreed that the Pre‐Reversal Enhancement (PRE) plasma vertical drift velocity plays a key role in their occurrence. In this study, it is examined how F‐layer meridional neutral winds affect the post‐sunset equatorial ionospheric irregularities using GOCE satellite data in the years 2012–2013 employing statistical analysis. The Rate of Total Electron content (TEC) Index (ROTI) is commonly used to monitor ionospheric irregularities. The finding reveals a strong correlation between ROTI values and the differences in meridional wind speeds at the north and south equatorial ionization anomaly (EIA) crests. Observations show that smaller speed differences (less than ∼5 m/s) between the north and south EIA crests support the formation of post‐sunset ionospheric irregularities, while larger speed differences (more than 5 m/s) do not support irregularity formation.


of NOAA SWPC alerts and warnings from 1 May 2024 to 16 May 2024.
Time series of indices, aa (as calculated by British Geological Survey) in yellow, aa (as calculated by International Service of Geomagnetic Indices) in blue, Dst in orange, ap in green, ap30 in red, Ap in purple, rAp (described in Section 2.2.7) in brown, SMR in pink and SYM‐H in gray. The negative of Dst, SMR and SYM‐H are plotted for visualization purposes.
Minutely (negative) SMR values (blue) and hourly averaged values, whilst the peak value is at 2235UT May 10, it is short lived and so the peak value in the hourly resolution data takes place at 02UT on May 11.
Plot of Kp index from 5 May 2024 to 11 May 2024.
Probability‐Probability (P–P) plots for the (a) ap, (b) ap30, (c) ap60, and (d) rAp extreme value theory model fits.
The Probability of the May 2024 Geomagnetic Superstorm

January 2025

·

162 Reads

In May 2024, a series of coronal mass ejections resulted in the first “severe” (G4‐level) geomagnetic storm watch in nearly 20 years. This event evolved into a significant space weather event, including an “extreme” (G5) geomagnetic storm, moderate (S2) solar radiation storm, and strong (R3) radio blackout. The widespread visibility of auroras at unusually low latitudes attracted global media attention. Using extreme value theory (EVT), this study estimates the return periods for the May 2024 storm based on several geomagnetic indices. The results indicate that while the storm's magnitude was a 1‐in‐12.5‐year event, its duration was a 1‐in‐41‐year event. This discrepancy highlights the storm's unusual longevity compared to its intensity. Updated EVT analyses incorporating recent data refine these return period estimates, providing critical insights into the frequency of such extreme space weather events.


TIE‐GCM ROPE ‐ Dimensionality Reduction: Part I

January 2025

·

39 Reads

Physics‐based models of the ionosphere‐thermosphere system have been touted as the next big thing in the context of drag modeling and space operations for decades. However, the computational complexity of such models have primarily kept them being used operationally. We recently demonstrated a proof‐of‐concept for developing what we call a reduced order probabilistic emulator (ROPE) for the thermosphere using the thermosphere ionosphere electrodynamics ‐ general circulation model (TIE‐GCM). The methodology uses a page out of dynamical systems theory to first reduce the order of the state using dimensionality reduction and then modeling the temporal dynamics in the reduced state space. The methodology uses an ensemble of temporal dynamic models to provide uncertainty estimates in the prediction. This work focuses on the dimensionality reduction step of the ROPE development process and addresses three limitations of the proof‐of‐concept: (a) extending the altitude upper boundary from 450 km to nearly 1000 km, (b) employing deep learning for nonlinear dimensionality reduction over principal component analysis (PCA) for improved performance during storm periods, and (c) maintaining the spatial resolution of the physical TIE‐GCM model, without down‐sampling, to preserve the spatial scales and variations. Results show overall performance boost over PCA for the high‐resolution and extrapolated data set as well as reduced reconstruction errors during storm‐time conditions. This work represents a major step toward operationalization.


Group picture from the IMC‐IV Workshop 2024 at GFZ‐Potsdam (Photo by A. Jordan).
Meeting Report: International Magnetosphere Coupling IV (IMC‐IV) Workshop GFZ‐Potsdam, Germany, 2–7 June 2024

January 2025

·

34 Reads

The International Magnetospheric Coupling (IMC‐IV) Workshop was held from 2–7 June 2024, at GFZ‐Potsdam, Germany and convened over 140 scientists to explore the solar wind and magnetosphere‐ionosphere‐thermosphere coupling. The workshop included eight sessions on topics ranging from solar wind, planetary magnetospheres, wave‐particle interactions to the ionosphere and upper atmosphere, and applications for space weather. Key subjects such as the need for advanced modeling, new satellite missions for validation, and the importance of understanding small and large‐scale space phenomena were emphasized by the participants. Two topical discussions focused on space weather awareness in Germany, and the future of Heliophysics research in Europe. The workshop concluded with recommendations for improved international mission coordination, better scientific communication, and enhanced public outreach through visualization tools. This meeting and the report pave the way for the formation of the European Heliophysics Community.


Journal metrics


3.8 (2023)

Journal Impact Factor™


58%

Acceptance rate


5.9 (2023)

CiteScore™


42 days

Submission to first decision


0.9 (2023)

Immediacy Index


0.00610 (2023)

Eigenfactor®


$2,890 / £2,190 / €2,540

Article processing charge

Editors