Geomagnetic Activity Research Articles

Unveiling the Core Patterns of High-Latitude Electron Density Distribution at Swarm Altitude

The ionosphere has distinctive characteristics under different solar and geomagnetic conditions, as well as throughout the seasons, and has a direct impact on our technological life in terms of radio communication and satellite navigation systems. In the pursuit of developing highly accurate ionospheric models and/or improving existing ones, understanding the various physical mechanisms that influence electron density dynamics is critical. In this study, we apply the Multivariate Empirical Mode Decomposition (MEMD) method to the electron density distribution in the mid-to-high latitude (above 50° magnetic latitude) regions in order to identify the dominant scales at which these mechanisms operate. The data were collected by the Swarm mission in the Northern Hemisphere. MEMD allows us to separate the main intrinsic modes and assess their relative contributions to the original one, thereby identifying the most important modes and the spatial scales at which they exert influence. Our study spanned the period from 1 January 2016 to 31 December 2021, which was characterized by low solar activity levels. This choice allowed for a more focused investigation of other variables influencing electron density distribution under similar solar activity conditions. We specifically examined the variations of the resulting modes in relation to different seasons and geomagnetic activity conditions, providing valuable insights into the complex behavior of the ionosphere in response to various external factors.

Properties of the Variability of the Maximum Density of the F2-Layer over Almaty at Different Levels of Solar and Geomagnetic Activity

The properties of the variability of the maximum density of the F2-layer Nm at different levels ofsolar and geomagnetic activity have been analyzed based on hourly data of the Almaty station (43.2° N, 104° E)for 1958–1988. The standard deviation σ(x) of the fluctuations of Nm relative to the quiet level(x = (Nm/Nm0 – 1) × 100, %) and the average shift of these fluctuations xave are used to characterize thisvariability. In this path, an empirical model of the F2-layer maximum density Nm0 for low geomagnetic activityhas been created. It has been found that the variability of Nm depends weakly on the level of solar activity.The dependence of the variability of Nm on geomagnetic activity is one of the main ones, along with thedependences of this variability on time of day and season. In general, the variance σ2(x) is smaller for quietconditions than for periods of high geomagnetic activity. However, during periods of high geomagnetic activity,a further increase in geomagnetic activity does not lead to an increase in the variance σ2(x). The saturationin the increase in the variance σ2(x) against the background of a continuing increase in geomagnetic activityand the absence of this saturation for the average shift xave seems to be a stable property of the variability ofthe mid-latitude ionosphere during periods of geomagnetic storms. This conclusion is based on an additionalanalysis of ionospheric variability according to data from the Irkutsk and Yamagawa stations, which arelocated about 10 degrees north and south of Almaty station, respectively.

Prediction of Proton Pressure in the Outer Part of the Inner Magnetosphere Using Machine Learning

AbstractThe information on plasma pressure in the outer part of the inner magnetosphere is important for simulations of the inner magnetosphere and a better understanding of its dynamics. Based on 17‐year observations from both Cluster Ion Spectrometry and Research with Adaptive Particle Imaging Detector instruments onboard the Cluster mission, we used machine‐learning‐based models to predict proton plasma pressure at energies from ∼40 eV to 4 MeV in the outer part of the inner magnetosphere ( = 5–9). Proton pressure distributions are assumed to be isotropic. The location in the magnetosphere, the property of stably trapped particles, and parameters of solar, solar wind, and geomagnetic activity from the OMNI database are used as predictors. We trained several different machine‐learning‐based models and compared their performances with observations. The results demonstrate that the Extra‐Trees Regressor has the best predicting performance. The Spearman correlation between the observations and predictions by the model is about 70%. The most important parameter for predicting proton pressure in our model is the value, which relates to the property of stably trapped particles. The most important predictor of solar and geomagnetic activity is F10.7 index. Based on the observations and predictions by our model, we find that no matter under quiet or disturbed geomagnetic conditions, both the dusk‐dawn asymmetry at the dayside with higher pressure at the duskside and the day‐night asymmetry with higher pressure at the nightside occur. Our results have direct practical applications, for instance, inputs for simulations of the inner magnetosphere or the reconstruction of the 3‐D magnetospheric electric current system based on the magnetostatic equilibrium.

Исследование влияния геомагнитной активности на функционирование систем железнодорожной автоматики в Арктической зоне России

The authors study the possible influence of geomagnetic activity of railway automatics in the Russian Arctic. For the analysis they use the indices of global and regional geomagnetic activity along with the archive of failures on the Northern section of the Oktyabrskaya railway in 2001—2006. The geomagnetic activity is found to be higher for days with failures than for days without the ones. This effect manifests itself in different types of geomagnetic disturbances, including magnetic storms, characterized by the storm index Dst, auroral activations, quantified in AE and EI indices, as well as in local index WY indicating spectral power of geomagnetic variations in milliHerz frequency range. The maximum differences for days with and without failures are obtained for index values averaged over 2—4 days. At the same time, the researchers have found no significant differences between failures for which a cause not related to geomagnetic disturbance is a priori indicated, and those for which such a cause is not identified.

The Global Variation of Low Ionosphere Under Action of Energetic Electron Precipitation

AbstractThe density in the D region is hard to observe on a large scale, because the density in this region is too low. In addition to solar radiation changes with a sinusoidal period of 24 hr, the energetic electron precipitation (EEP) is the major source for the ionospheric D region. In this study, we utilized observations from NOAA/MetOp satellites to derive the electron density profile from 60 to 150 km under the influence of EEP. It seems that the derived electron profile is roughly consistent with the observed one. Additionally, using above derived method of low ionospheric (D and E regions) electron profile based on the observations of NOAA/MetOp satellites, the actions of EEP on global ionosphere under different level of geomagnetic activity are analyzed. We find that EEP has a significant impact on the electron density in the altitude range from 60 to 90 km in the L ∼ 4.5–8 and MLT ∼ 9–14 regions, particularly during strong geomagnetic activity. A peak in Total Electron Content (TEC) was observed at MLT ∼ 11–12 and L ∼ 5.5–6, exhibiting an enhancement approximately 19.96 times higher than that observed during quiet periods. However, the effects of EEP decrease as altitude increases. The TEC enhancement at altitudes from 90 to 120 km during strong geomagnetic activity is like that at 60–90 km, but only about 2.08 times higher than that during quiet periods. In addition, only a weak increase is detected around midnight at altitudes from 120 to 150 km.

A Nonlinear Dependence on the Geomagnetic Activity of the Ratio of the Maximum Stream Flux of Charged Particles in a Geostationary Orbit to the Minimum Stream Flux

A new mathematical model was proposed using an ordinary differential equation that analytically(when the index of geomagnetic activity Kp = const or Kp ≈ const) or numerically (if Kp(t) ≠ const) describesperpendicular (for a pitch angle of 90°) differential or integral fluxes of relativistic electrons in a geostationary(geosynchronous) orbit, as well as in any circular orbit in the Earth’s magnetosphere. The model assumes thatthe fluxes depend on the local time LT in orbit, the Kp, MacIlvwaine parameter and L, and the perpendiculardifferential flux or integral flux of relativistic electrons taken at 0000:00 LT. We use observations of relativistic(2 MeV) electron fluxes averaged over the local time hour along the orbit of the GOES spacecraft from 1995to 2009. The model is compared with these data. Almost perfect agreement was obtained for observationswith the model, where the prediction efficiency of predicting the accuracy of the model at PE = 0.9989. Usingsimilar data from the GOES 10 allows one to obtain PE = 0.9924. The proposed formulas make it possible tofind, for example, the average value of the perpendicular integral flux of relativistic electrons per day and topredict the maximum perpendicular integral flux of relativistic electrons in the geostationary orbit approximately1 day ahead. The nonlinear effect is theoretically predicted in the form of a nonlinear dependence ofthe ratio of the maximum perpendicular integral flux to the minimum flux of charged particles in the geostationaryorbit from the Kp-index of geomagnetic activity. Thus far, comparison of the model has been madewith the averaged integral relativistic electron flows fluxes produced for the 0 ≤ Kp 6 range with a predictedmaximum flow flux ratio of 24.4139 times at Kp = 8 and with the prediction efficiency of predicting the accuracyof the nonlinear effect PE = 0.8678.

Polar Cap Patches Scaling Properties: Insights from Swarm Data

Among the effects of space weather, the degradation of air traffic communications and satellite-based navigation systems are the most notable. For this reason, it is of uttermost importance to understand the nature and origin of ionospheric irregularities that are at the base of the observed communication outages. Here we focus on polar cap patches (PCPs) that constitute a special class of ionospheric irregularities observed at very high latitudes in the F region. To this purpose we use the so-called PCP flag, a Swarm Level 2 product, that allows for identifying PCPs. We relate the presence of PCPs to the values of the first- and second-order scaling exponents and intermittency estimated from Swarm A electron density fluctuations and to the values of the Rate Of change of electron Density Index (RODI) for two different levels of geomagnetic activity, over a time span of approximately 3.5 years starting on 16 July 2014. Our findings show that values of RODI, first- and second-order scaling exponents and intermittency corresponding to measurements taken inside PCPs differ from those corresponding to measurements taken outside PCPs. Additionally, the values of the first- and second-order scaling exponents and of intermittency indicate that PCPs are in a turbulent state. Investigation of the coincidence of loss of lock (LoL) events with PCPs displayed that approximately 57.4% of LoLs in the Northern hemisphere and 45.7% in the Southern hemisphere occur in coincidence of PCPs when disturbed geomagnetic activity is considered. During quiet geomagnetic conditions these percentages decrease to 51.4% in the Northern hemisphere and to 20.1% in the Southern hemisphere.

Annual Variations in the Near-Earth Solar Wind

Earth’s orbit and rotation produces systematic variations in geomagnetic activity, most notably via the changing orientation of the dayside magnetospheric magnetic field with respect to the heliospheric magnetic field (HMF). Aside from these geometric effects, it is generally assumed that the solar wind in near-Earth is uniformly sampled. But systematic changes in the intrinsic solar wind conditions in near-Earth space could arise due to the annual variations in Earth heliocentric distance and heliographic latitude. In this study, we use 24 years of Advanced Composition Explorer data to investigate the annual variations in the scalar properties of the solar wind, namely the solar wind proton density, the radial solar wind speed and the HMF intensity. All parameters do show some degree of systematic annual variation, with amplitudes of around 10 to 20%. For HMF intensity, the variation is in phase with the Earth’s heliocentric distance variation, and scaling observations for distance largely explains the observed variation. For proton density and solar wind speed, however, the phase of the annual variation is inconsistent with Earth’s heliocentric distance. Instead, we attribute the variations in speed and density to Earth’s heliographic latitude variation and systematic sampling of higher speed solar wind at higher latitudes. Indeed, these annual variations are most strongly ordered at solar minimum. Conversely, combining scalar solar wind parameters to produce estimates of dynamic pressure and potential power input to the magnetosphere results in solar maximum exhibiting a greater annual variation, with an amplitude of around 40%. This suggests Earth’s position in the heliosphere makes a significant contribution to annual variations in space weather, in addition to the already well-studied geometric effects.

Properties of the Variability of the Maximum Density of the F 2-Layer over Almaty at Different Levels of Solar and Geomagnetic Activity

Properties of the Variability of the Maximum Density of the F 2-Layer over Almaty at Different Levels of Solar and Geomagnetic Activity

Wavelike Oscillations in High Latitude Thermospheric Doppler Temperature and Line‐Of‐Sight Wind Observed Using All‐Sky Imaging Fabry‐Perot Spectrometers

AbstractMultiple years of thermospheric wind and temperature data were examined to study gravity waves in Earth's thermosphere. Winds and temperatures were measured using all‐sky imaging optical Doppler spectrometers deployed at two sites in Alaska, and three in Antarctica. For all sites, oscillatory perturbations were clearly present in high‐pass temporally filtered F‐region line‐of‐sight (LOS) winds for the majority of the clear‐sky nights. Oscillations were also discernible in E‐region LOS wind and F‐region Doppler temperature, albeit less frequently. Oscillation amplitudes correlated strongly with auroral and geomagnetic activity. Observed wave signatures also correlated strongly between geographically nearby observing sites. Amplitudes of LOS wind oscillations were usually small when viewed in the zenith and increased approximately with the sine of the zenith angle—as expected if the underlying motion is predominantly horizontal. Scanning Doppler Imager instruments observe in many look directions simultaneously. Phase relationships between perturbations observed in different look directions were used to identify time intervals when the oscillations were likely to be due to traveling waves. However, a number of instances were noted in which the oscillations had characteristics suggesting geophysical mechanisms other than traveling waves—a recognition that was only possible because of the large number of look directions. Lomb‐Scargle analysis was used on a representative subset of days to resolve the spectral distributions of the wind and temperature oscillations. F‐region wind oscillations on days analyzed this way exhibited periods typically ranging from 60 min and above. By contrast, E‐region wind oscillation periods were as short as 30 min.

Parametrization of Energetic Ion and Electron Fluxes in the Near‐Earth Magnetotail

AbstractThe near‐Earth plasma sheet region is the main source of energetic (tens to hundreds keV) ion and electron populations transported by convection and injections into the inner magnetosphere. Energetic ions from the plasma sheet contribute to the ring current, whereas energetic electrons contribute to the radiation belt seed population for further acceleration to relativistic energies. Near‐Earth plasma sheet energetic fluxes have been traditionally used to set boundary conditions for radiation belt and ring current models. This study provides an empirical parametrization for ∼75 keV flux intensity as a function of the geomagnetic activity index auroral electrojet and the equatorial magnetic field Bz. Such parametrization includes the dynamic magnetic field configuration in the near‐Earth plasma sheet and may be merged with empirical magnetic field models. We also provide models extending this parametrization to the [20, 300] keV of electron energy range and [75, 300] keV of ion energy range. The parametrization is developed based on THEMIS and Geostationary Operational Environmental Satellite measurements, and verified by comparison with MMS measurements in the near‐Earth plasma sheet. This parametrization incorporates meso‐scale transient flux variations associated with Bz perturbations into ring current and radiation belt simulations.

A Nonlinear Dependence on the Geomagnetic Activity of the Ratio of the Maximum Flux of Charged Particles in a Geostationary Orbit to the Minimum Flux

A Nonlinear Dependence on the Geomagnetic Activity of the Ratio of the Maximum Flux of Charged Particles in a Geostationary Orbit to the Minimum Flux

Quantify the Information Contributions of External Drivers to Global Ionospheric TEC

AbstractThe ionospheric variabilities are influenced by various drivers, and quantifying their relative contributions is beneficial to ionospheric modeling and ionospheric forecasting, but this is still a significant challenge in the ionospheric community. In present paper, the global ionosphere maps, solar indices, solar wind and interplanetary parameters, geomagnetic indices, and lower atmospheric parameters from 2001 to 2021 are collected, and the Liang‐Kleeman information flow is used to measure the amount of information exchanges from external drivers to ionospheric TEC at different geographical regions, and the normalized Liang‐Kleeman information flow is adopted to quantify their relative information contributions. The results show that the geographical distribution of information flow has prominent hemispheric asymmetry, and overall, the information flows from solar indices (solar wind and interplanetary parameters, geomagnetic indices) to ionospheric TEC in the northern hemisphere are larger than those in the southern hemisphere. Ionospheric self‐influence (i.e., internal processes of the ionosphere) has the largest information contribution to ionospheric TEC, accounting for about 46.78%. Solar activity manifested as P index has the second largest contribution (nearly 35.29%). Solar wind and interplanetary activity represented as IMF magnitude contributes around 2.94%, geomagnetic activity characterized as ASYH index accounts for about 1.52%, lower atmospheric forcing contributes nearly 7.75%, and noise accounts for approximately 5.73%. What's more, the performances of the IRI‐2020 model are evaluated from the perspective of information flow.

Assessing the performance of IRI-2016 and IRI-2020 models using COSMIC-2 GNSS radio occultation TEC data under different magnetic activities over Egypt

Abstract Variations in Total Electron Content (TEC) between the COSMIC-2, IRI-2016, and IRI-2020 are considered under different levels of geomagnetic storm activity: minor, moderate, and severe. TEC values are scrutinized at three levels of the Kp index, which serves as a metric for gauging the strength of a magnetic storm (Kp = 3.0, Kp = 6.0, Kp = 8.0) and across four-time intervals throughout 24 h to understand the performance of the models during both day and night-time conditions. Statistical analysis reveals that the standard deviation of TEC variations is lower during minor storms than moderate and severe. The comparison of variations between COSMIC-2 Radio Occultation TEC and both IRI-2016 and IRI-2020 models revealed more substantial discrepancies during day-time intervals; This was likely attributed to the dynamic and complex nature of the ionosphere influenced by solar radiation and other factors. Comparative analysis across the three levels of storm activity demonstrated that IRI2020 provided improved results over IRI2016, particularly during minor geomagnetic storm events. The study demonstrates that IRI2020 is more accurate than IRI-2016 at forecasting ionospheric conditions, especially at night and during moderate geomagnetic storm activity periods. Both models, however, provide valuable insights during challenging space weather conditions, and the results demonstrate their utility in understanding and forecasting the ionosphere’s behavior. The results yield valuable insights into space weather conditions and their effects on technology and communication, highlighting the potential for further improvement in TEC prediction models.

Possible Influence of Solar Cyclicity on Extratropical Cyclone Trajectories in the North Atlantic

In this work, we study long-term changes in the main directions of movement (storm tracks) of extratropical cyclones in the North Atlantic for the period 1873–2021, based on the data of the MSLP (Mean Sea Level Pressure) archives from the Climatic Research Unit and NCEP/DOE AMIP-II reanalysis. It was found that in the cold half of the year, the mean latitudes of storm tracks undergo oscillations with the periods ~80–100, ~40–45, and ~22 years, which indicates their possible association with solar cyclicity. Cyclone trajectories were found to shift northward at the minimum of the Gleissberg secular cycle (~1900–1930) and southward at its maximum (~1940–1960). The secular variations are the most pronounced in the western part of the North Atlantic, with the peak-to-peak amplitude reaching ~3–5°, and disappear at longitudes east of Greenwich. On the bidecadal timescale, cyclone tracks were found to shift noticeably to the north in even solar cycles and weakly to the south in odd ones. The most significant northward shifts (~1–2°) were detected during the descending phase and the minimum of the solar cycle (from the second to the sixth year after the solar maximum) in the eastern part of the North Atlantic (longitudes 30–10° W). The detected oscillations of cyclone trajectories may be caused by long-term changes in the intensity of the stratospheric polar vortex associated with galactic cosmic ray variations and geomagnetic activity.

Equatorial plasma bubbles features over the Brazilian sector according to the solar cycle and geomagnetic activity level

Equatorial plasma bubbles (EPBs) can lead to signal degradation, affecting the measurement accuracy. Studying EPBs and their characteristics has gained increasing importance. The characteristics of EPBs were investigated using the rate of total electron content (TEC) index (ROTI) maps under different solar and magnetic activity conditions during two periods: July 2014–July 2015 (solar maximum activity with F10.7: 145.9 × 10−22 W⋅m−2⋅Hz−1) and July 2019–July 2020 (solar minimum activity with F10.7: 69.7 × 10−22 W⋅m−2⋅Hz−1). We also divided this analysis according to the magnetic activity levels based on Kp and Dst (disturbance storm time) indices, classified as follows: quiet+ (Kp ≤3 and Dst >−30 nT), quiet− (Kp ≤3 and Dst <−30 nT), disturbed weak (−50 nT <Dst ≤−30 nT), moderate (−100 nT <Dst ≤−50 nT), and intense (Dst ≤−100 nT). The ROTI is calculated using the slant TEC with the carrier phase, and its keograms are used to extract the zonal velocity and distance. Our statistical investigation shows the occurrence rate, duration, zonal drift velocity, and inter-bubble zonal distance of EPBs over the Brazilian sector. The latitudinal extension and zonal drift velocity of EPBs are higher during the solar maximum than those in the solar minimum. In addition, EPBs are found with unusually long durations, remaining until the morning (∼12 UT), and 10% of EPB observations occurred on the winter solstice.

MOVEMENT ACTIVITY OF ROUND GOBY (NEOGOBIUS MELANOSTOMUS PALLAS, 1814) UNDER THE INFLUENCE OF GEOMAGNETIC FIELDS

Introduction. Behavioral traits of fish are an important manifestation of their response to external stimuli. Locomotor activity reflects the influence of specific factors on fish behavior and the role they play in their lives. Among the least studied factors are fluctuations in the geomagnetic field and their impact on the locomotor activity of the round goby Neogobius melanostomus Pallas, 1814. Aim. The aim of this study was to investigate the influence of fluctuations in the geomagnetic field on the locomotor activity of the commercially important speciesof benthic ichthyofauna, the round goby Neogobius melanostomus Pallas, 1814, in the Black Sea. Methods. The ichthyological material was collected in the coastal waters of the Odesa Bay, ranging from Cape Northern Odesa to Cape Velikyi Fontan, during a scientific research survey conducted by the I.I. Mechnikov Odesa National University (ONU). Laboratory investigations were conducted in the aquarium room of the Aquatic Biology and General Ecology Department of the I.I. Mechnikov Odesa National University. Hourly changes in fish locomotor activity were recorded and analyzed as a result of the study. During the specified observation period, periods of increased geomagnetic activity were identified and compared to periods when this activity was within normal range. An original methodology called “Computer Vision Method” was employed for processing the visual data and tracking of the laboratory animals. Results. The maximum values of geomagnetic activity during the observation period, classified as a first intensity magnetic storm of class G1, were observed on February 3, 2022, from 6 to 12 hours, and on February 4 and 10, 2022, from 15 to 21 hours. When comparing the levels of fish locomotor activity during these time intervals with their activity during the time intervals on days with minimal geomagnetic activity (February 8, 9, and 15, 2022), a decrease in locomotor activity of individuals during magnetic storms of 27% to 42% can be observed. Conclusion. The fluctuations in locomotor activity of the round goby during increased geomagnetic activity up to the level of magnetic storms indicate the ability of this species to perceive magnetic fields and respond to changes in its properties. Specifically, when the Kp-index values corresponded to a first intensity geomagnetic storm (G1), a decrease in fish locomotor activity was observed compared to the activity during the same time intervals on days with minimal geomagnetic activity.

Variation of Total Electron Content over Nepal during Geomagnetic Storms: GPS Observations

Geomagnetic storms have very profound effects on the Total Electron Content (TEC) of the ionosphere. In order to investigate the equatorial and low-latitude ionospheric response to the geomagnetic storms of varying intensities, a detailed study of vertical TEC (VTEC) variations resulting from Global Positioning System (GPS) data acquired at four GPS stations in Nepal along 80°–90° E longitude and 26°–30° N latitude sector has been carried out in the present work. The results were analyzed with other favorable inducing factors (solar wind parameters and geomagnetic indices) affecting TEC to constrain the causative factor. Positive phases are observed for all the storms studied. During the severe geomagnetic activity, the deviation was ~18 TECU, while it was recorded ~12 TECU and ~8 TECU during moderate and minor geomagnetic activity, respectively. The Detrended Cross-Correlation Analysis (DXA) illustrates that the value of the hourly average VTEC of the BESI station was found to have a strong positive correlation with other stations in all types of storm events, indicating a similar response of all stations towards the space weather events. In addition, the correlation of VTEC with solar wind parameters and geomagnetic indices illustrated that the VTEC shows a strong positive association with solar wind velocity (Vsw) in all three geomagnetic events. In contrast, the correlation of plasma density (Nsw), interplanetary magnetic field (IMF-Bz), the symmetric horizontal component of geomagnetic field (SYM-H), and Geomagnetic Auroral Electrojet (AE) index with VTEC vary with the intensity of the storm. Overall results of the study have revealed the characteristic features of TEC variation over Nepal regions during magnetic storms, which validates earlier research on ionospheric responses to geomagnetic storms and theoretical assumptions.

Correlation Between COSMIC Rayn Intensity and Geomagnetic Activity Indices

This study conducts an extensive analysis to explore the correlation between the monthly average cosmic ray intensity (CRI) and indicators of geomagnetic activity, primarily focusing on the Ap and Kp indices. The study covers two distinct time intervals, 1996-2008 and 2009-2019. By analyzing the correlation between CRI and geomagnetic indices, we constructed linear graphs depicting their associations. The results consistently reveal a noticeable inverse correlation between monthly average CRI and the Ap index. Correlation coefficients were computed to quantify these relationships using established statistical methods. Between 1996 and 2008, a strong negative correlation of -0.89 was observed between CRI and the Ap index. Similarly, during 2009-2019, a significant negative correlation of -0.39 was noted. The monthly average CRI exhibited a negative correlation of -0.72 with the Kp index during 1996-2008, and a more pronounced negative correlation of -0.70 during 2009-2019. These findings underscore the intricate interplay between cosmic ray intensity and geomagnetic activity, offering insights into the broader understanding of space weather phenomenon

Equatorial spread-F forecasting model with local factors using the long short-term memory network

The predictability of the nighttime equatorial spread-F (ESF) occurrences is essential to the ionospheric disturbance warning system. In this work, we propose ESF forecasting models using two deep learning techniques: artificial neural network (ANN) and long short-term memory (LSTM). The ANN and LSTM models are trained with the ionogram data from equinoctial months in 2008 to 2018 at Chumphon station (CPN), Thailand near the magnetic equator, where the ESF onset typically occurs, and they are tested with the ionogram data from 2019. These models are trained especially with new local input parameters such as vertical drift velocity of the F-layer height (Vd) and atmospheric gravity waves (AGW) collected at CPN station together with global parameters of solar and geomagnetic activity. We analyze the ESF forecasting models in terms of monthly probability, daily probability and occurrence, and diurnal predictions. The proposed LSTM model can achieve the 85.4% accuracy when the local parameters: Vd and AGW are utilized. The LSTM model outperforms the ANN, particularly in February, March, April, and October. The results show that the AGW parameter plays a significant role in improvements of the LSTM model during post-midnight. When compared to the IRI-2016 model, the proposed LSTM model can provide lower discrepancies from observational data.Graphical