Driver of the Positive Ionospheric Storm over the South American Sector during 4 November 2021 Geomagnetic Storm

Abstract

During geomagnetic storms, ionospheric storms can be driven by several mechanisms. Observations performed using ground- and space-based instruments were used to reveal the driver of the positive ionospheric storm over the South American sector during the 4 November 2021 geomagnetic storm. The positive storm appeared from 10:30 UT to 18:00 UT and covered the region from 40°S to 20°N. The maximum magnitudes of TEC (Total Electron Content) enhancement and relative TEC enhancement were about 20 TECU and 100%, respectively. Defense Meteorological Satellite Program (DMSP) also observed a significant electron density increase over South America and the eastern Pacific Ocean. In the meantime, about 50% ∑O/N₂ enhancement was observed by the Global-scale Observations of the Limb and Disk (GOLD) satellite at low latitudes. Ionosonde observations (AS00Q and CAJ2M) registered an ~80 km uplift in F2 peak height (HmF2) and a prominent F2 peak electron density (NmF2) increase ~3 h after the uplift. A prominent enhancement in the cross-polar cap potential (CPCP) in the southern hemisphere was also observed by Super Dual Auroral Radar Network (SuperDARN) one hour earlier than the HmF2 uplift. Measurements of the Ionospheric Connection Explorer satellite (ICON) showed that the outward E×B drift was enhanced significantly and that the horizontal ion drift was poleward. According to the ICON ion drift observations, the HmF2 uplift was caused by an electric field rather than equatorward neutral wind. We propose that the enhanced eastward electric field dominated the positive ionospheric storm and that the thermospheric composition variation may have also contributed.

1. Introduction

The ionospheric response to geomagnetic storms has been studied extensively due to its important role in the magnetosphere–ionosphere–thermosphere (MIT) system and the impacts on radio communications, satellite navigations, and spacecrafts [1,2,3,4,5,6]. The variations in the ionosphere can be significant when high-speed solar wind is accompanied by a long-time, southward interplanetary magnetic field (IMF) component [7]. Compared with quiet time, the ionospheric electron density variations during geomagnetic storms can be positive and negative, which are referred to as “positive” and “negative” ionospheric storms, respectively. Positive ionospheric storms can cause significant ionospheric delay and scintillations, while negative ionospheric storms can induce severe blackouts in Global Navigation Satellite Systems (GNSSs) and high-frequency (HF) radio communications [5,8,9,10].

Electric fields, thermospheric composition, and neutral winds can all drive the ionospheric response to geomagnetic storms at low latitudes [11,12,13,14,15]. The southward turning of the interplanetary magnetic field (IMF) Bz component enhances the high-latitude convection rapidly, which can penetrate into low latitudes since the magnetosphere cannot respond to the orientation variation of IMF Bz in time [16,17,18,19,20]. The prompt penetration electric fields (PPEFs) are eastward on the dayside and westward on the nightside. On the dayside, the expanded eastward electric field causes plasma vertical drift with the geomagnetic field, and the lifted plasma accumulates due to the lower recombination rate at high altitudes and generates electron density and TEC enhancements [21,22]. Using the multi-instrument data from a chain at magnetic conjugate locations in the 100°E sector, Kalita et al. [23] investigated the ionospheric response during the St. Patrick’s Day storms in 2013 and 2015; they suggested that the direction of Bz was essential to the evening equatorial electrodynamics, and the storm-induced upward vertical drift at sunrise was found to be larger than the drift during evening prereversal enhancement (PRE). Lu et al. [24] simulated the ionospheric and thermospheric variations during a PPEF event using Thermosphere–Ionosphere–Mesosphere Electrodynamic General Circulation Model (TIMEGCM). The results showed that electric fields were the dominant drivers of vertical ion drift at low latitudes, while vertical disturbance meridional winds exceeded the electric field effect. During the main phase of the 17 March 2015 geomagnetic storm, the enhanced zonal electric field resulted in a rapid HmF2 uplift and a strong positive ionospheric storm effect [25].

In addition to electric fields, thermospheric composition variations can also affect the ionosphere. During geomagnetic storms, the Joule and participate heating induced by enhanced energy deposition from the magnetosphere triggers equatorward wind surges. On the one hand, equatorward wind pushes plasma upward along geomagnetic lines and causes a positive effect on electron density due to the reduced recombination rate with the increase in altitude [26]. Millstone Hill and Arecibo incoherent scatter radars registered a strong positive storm on 10 September 2005. Based on the simulations of TIEGCM, Lu et al. [27] suggested that the positive storm was primarily caused by enhanced meridional wind rather than the penetration magnetospheric electric field.

On the other hand, downward neutral wind induced by Joule and participate heating at high latitudes tends to bring O-atom-rich air to lower altitudes and increase the O/N₂ ratio at low latitudes. The increase in the O/N₂ ratio in the thermosphere can lead to an increase in the O⁺ production rate and a decrease in the recombination rate, which further cause TEC and electron density enhancement at middle and low latitudes [28,29]. At the end of the main phase of the 22–23 June 2015 geomagnetic storm, no significant vertical plasma motion was observed, but electron density and TEC increased drastically. Astafyeva et al. [16] suggested that the effect of PPEFs was offset by disturbance dynamo electric fields (DDEFs) and that the positive storm was largely influenced by thermospheric compositions. Younas et al. [30] conducted statistical calculations on the relationship between ∑O/N₂ and TEC at low latitudes during four intense storms of solar cycle 24 and found that a positive storm effect showed good consistency with ∑O/N₂ enhancement.

Since the ionospheric response to geomagnetic storms is an essential part of solar wind–magnetosphere–ionosphere–thermosphere coupling, it is of great importance to distinguish the roles played by electric fields, neutral winds, and thermospheric composition in generating ionospheric storms. However, the above-mentioned drivers could affect the ionosphere simultaneously during geomagnetic storms. Due to the lack of neutral wind and electric field observations, it is not easy to determine the dominant factor, especially for PPEFs and equatorward winds, since they can both generate HmF2 uplifts and positive storms.

In this study, we focused on the positive ionospheric storm over the South American sector on 4 November 2021. GNSS TEC, DMSP in situ electron density, and ionosonde observations were employed to analyze the characters of this positive storm. The drivers of this event were investigated based on GOLD ∑O/N₂, CPCP, and ICON horizontal and meridional ion drift observations.

2. Data and Methodology

The GNSS VTEC data used in this study were obtained from Massachusetts Institute of Technology (MIT)’s Haystack Observatory, which provides global VTEC maps with 5-minute temporal resolution and 1° × 1° spatial resolution in latitude and longitude.

∑O/N₂ data were derived from the observations obtained by the GOLD airglow imaging instrument, which is on board the SES-14 communication satellite in geostationary orbit at 47.5° west longitude. During the daytime, the GOLD imager scans the Earth every 30 min in the range from 120°W to 30°E longitude and from 70°S to 70°N latitude [31,32,33]. The horizontal resolution of ∑O/N₂ data used in this study was 1° × 1° (latitude and longitude).

The ionospheric F2 peak electron density (NmF2) and F2 peak height (HmF2) observations from AS00Q and CAJ2M ionosonde stations were used in this study. The ionosonde data were obtained from National Centers for Environmental Information (NCEI) via ftp service (ftp.ngdc.noaa.gov, accessed on 28 November 2022). NCEI, which is co-located with World Data Service for Geophysics, has assembled some 40,000 station-months of scaled digital Ionospheric vertical incidence parameters from about 130 sites worldwide and offers them on CD-ROM (Ionospheric Digital Database) for general distribution. The distribution of the two ionosonde stations is given in Figure 1.

Cross-polar cap potential (CPCP) data derived from Super Dual Auroral Radar Network (SuperDARN) were also used in this study to investigate the high-latitude electric field. SuperDARN is a high-frequency (HF) radar network that consists of 36 radars over the high and middle latitude regions in the northern hemisphere and southern hemisphere [34,35].

In situ electron density data were recorded by the “Special Sensor Ions, Electrons and Scintillation” (SSIES) instrument on board the Defense Meteorological Satellite Program (DMSP) F-17 satellite. The satellite flies in Sun-synchronous, near-polar orbits, at inclinations of about 98.8° at an altitude of about 850 km.

Horizontal ion velocity data were recorded by the Ion Velocity Meter (IVM) instrument on the Ionospheric Connection Explorer (ICON) satellite, which orbits the Earth at about 575 km at the inclination of 27° [36]. The IVM instrument is composed of an ion drift meter and a retarding potential analyzer to measure parameters including ion temperature, ion composition, ion density, and ion drift velocity [37].

3. Results

3.1. Geomagnetic Conditions

An intense geomagnetic storm occurred during 3–5 November 2021. Figure 2 shows the geospacer indices during 3–6 November: (a) IMF Bz component in Geocentric Solar Magnetospheric (GSM) coordinates; (b) symmetric ring current index (SYM-H); (c) east–west component of the interplanetary electric field (Ey) index; (d) asymmetric ring current (ASY-H) index; (e) Auroral electrojet (AE) index; (f) Kp index; (g) F10.7 index. Bz turned southward at about 20:30 UT on 3 November, and sudden storm commencement (SSC) occurred. SYM-H decreased from 21:00 UT on 3 November and reached its minimum of −117 nT on 4 November. Ey increased rapidly from 0 to 13 mV/m and oscillated with the Bz component. The two peaks in the ASY-H and AE indices around 22:00 UT on 3 November and 02:00 UT on 4 November coincided with downward SYM-H around those times. The subsequent long-time enhancement in the ASY-H and AE indices from ~04:00 UT to ~15:00 UT on 4 November suggests that there were strong partial ring current, PPEF, auroral activity, and joule heating, which might have induced equatorward neutral wind surges [38]. Kp increased from 4 to 6 and reached its maximum value of 7 at 12:45 UT on 4 November. The F10.7 index stayed at around 85 during 3–6 November.

On 3 November, until 20:30 UT, there was no significant disturbance in Bz, SYM-H, and AE indices, and the Kp index was smaller than 3. For analyzing the ionospheric response on 4 November over the South American sector, 3 November was selected as the quiet day. The TEC difference (DTEC; left column) and the relative TEC difference (RDTEC; right column) between the storm day (4 November) and the quiet day (3 November) were calculated, and the results are shown in Figure 3. As we can see in the left column of Figure 3, the positive ionospheric storm started at 11:00 UT in the east of the South American sector and expanded to the latitude range between 40°S and 20°N at 13:20 UT. From 13:20 UT to 15:40 UT, the magnitude of TEC enhancement further increased. At 18:00 UT, TEC enhancement decayed. In the right column, the magnitude of the RDTEC was about 80% between 20°S and 20°N and 40% between 40°S and 20°S at 13:20 UT. At 15:40 UT, the magnitude of TEC enhancement at the magnetic equator decreased to about 25%, while the magnitude in the southern EIA crest region increased to 50%. At 18:00 UT, the magnitude of the RDTEC over the whole South American sector decreased.

3.2. Electron Density Enhancement

Figure 4 shows the satellite orbit of DMSP (left column) and in situ electron density comparison (right column) between the storm day and the quiet day close to 11:05 UT and 12:45 UT. The satellite orbit and electron density values on the storm day and the quiet day are presented by red lines and black lines, respectively. At 11:05 UT, DMSP F-17 passed the north and west coasts of the South American sector. The electron density values between [20°S, 20°N] on the storm day at 11:05 UT were larger than those on the quiet day. At 12:45 UT, DMSP F-17 passed the eastern Pacific Ocean. With the intensification of the positive ionospheric storm, the electron density enhancement became more prominent. The electron density values on the storm day increased to three times of quiet-day values at 10°N and were significantly larger than the values on the quiet day at other latitudes.

3.3. O/N₂ Variation

Figure 5 gives the relative difference in GOLD ∑O/N₂ between the storm day and the quiet day from 10:22 UT to 20:22 UT. At 10:22 UT, there was ∑O/N₂ depletion with the magnitude of ~40% at middle and high latitudes over the eastern part of South America and the Atlantic. At the same time, ∑O/N₂ was enhanced at low latitudes and in the equatorial region at the magnitude of 30%. From 12:22 UT to 14:22 UT, GOLD ∑O/N₂ observations covered the most of South America, and the magnitude of ∑O/N₂ enhancement increased to ~50%. From 16:22 UT to 18:22UT, there were no obvious variations in the morphology of ∑O/N₂ enhancement, but its magnitude decreased to ~20%. At 20:22 UT, GOLD ∑O/N₂ observations covered the western part of South America, and the magnitude of ∑O/N₂ enhancement between 30°S and 10°N was ~35%, which might have been correlated with other events.

3.4. NmF2 and HmF2 Variations

The HmF2 and NmF2 comparisons between the storm day and the quiet day observed by the AS00Q and CAJ2M ionosonde stations are shown in Figure 6. As we can see in Figure 6a, HmF2 at AS00Q increased to about 400 km at 10:00 UT on 4 November, while it was about 320 km on 3 November. Then, HmF2 fell to 320 km at 12:00 UT, and there were no significant differences in HmF2 between the storm day and the quiet day at the AS00Q station after that time. In Figure 6c, NmF2 on the storm day was larger than that on the quiet day between 12:00 and 14:00 UT. In Figure 6b, HmF2 at CAJ2M on the storm day increased rapidly from 11:30 UT and reached its maximum value of 410 km at 12:30 UT; then, it decreased to a height similar to that of the quiet day. In Figure 6d, NmF2 enhancement started from about 14:00 UT and reached its maximum value of 24.5×10¹¹ el/m³ at 15:30 UT. NmF2 values between 15:00 and 17:00 UT on the storm day were prominently larger than those on the quiet day.

3.5. CPCP

Figure 7 shows the comparison of cross-polar cap potential (CPCP) values at the South Pole between the quiet day (black line) and the storm day (red line). The shaded area indicates the time period of the positive ionospheric storm. As we can see in Figure 7, the CPCP on the quiet day was less than 60 kV and was significantly smaller than that on the storm day most of the time. There were two peaks between 00:00 UT and 06:00 UT that were not correlated with the positive storm, since the South American sector was on the night side during that time. From ~08:45 UT on 4 November, the CPCP showed rapid growth and reached ~85 kV at 09:10 UT and oscillated around ~70 kV until ~16:00 UT, and decreased to ~30 kV at ~21:10 UT.

4. Discussion

During 11:00–18:30 UT on 4 November, significant TEC and electron density enhancement were observed by GNSS and DMSP. Three mechanisms may have been responsible for this positive ionospheric storm event: thermospheric composition changes, equatorward neutral winds, and PPEFs. During geomagnetic storms, enhanced ion drag and pressure gradient force caused by Joule heating can induce upward (downward) neutral wind and O/N₂ depletion (enhancement) at high (low) latitudes [29,39]. As shown in Figure 5, significant ∑O/N₂ occurred at low latitudes, which may have contributed to the positive storm.

However, an F2 peak height uplift was also observed at both ionosonde stations, and NmF2 increased 2–3 h after the F2 peak height uplift (Figure 6). Since PPEFs and equatorward neutral wind can both increase the F2 peak height and lead to TEC enhancement [24,26,40], it is not easy to the distinguish their roles in this positive storm based on the above observations.

For further analyzing the uplift of HmF2, the storm-day horizontal and meridional ion drifts observed by the ICON satellite were compared with those on the quiet day. Figure 8 shows ICON ion horizontal velocity on the quiet day (3 November; black arrows) and storm day (4 November; red arrows) around 10:40 UT (left column) and 12:20 UT (right column). The magnetic equator and magnetic latitudes of 15° on both sides are indicated by blue lines. The second row of the X-axis denotes the local time at which the ICON satellite passed the longitudes in the first row of X-axis. At 10:40 UT (left column), the horizontal ion velocity on 3 November was southward, between 60 and 100°W (local time 03:24–07:10), and was induced by the equatorward neutral wind on the nightside. Due to the northward neutral wind from the summer hemisphere to winter hemisphere, the horizontal ion velocity between 25 and 60°W (local time 07:10–09:10) was generally northward. On 4 November, the southward horizontal ion velocity in the geomagnetic northern hemisphere was enhanced by the storm-induced equatorward neutral wind [41]. In the geomagnetic southern hemisphere, it was northwestward and southeastward between [−15, 0] and [−30, −15] magnetic latitude, respectively. At 12:20 UT (right column), the horizontal ion drift on 3 November was similar to that at 10:40 UT. However, the horizontal ion drift in the geomagnetic southern hemisphere on 4 November was southeastward and generally perpendicular to the magnetic equator.

The comparison of ICON meridional ion drifts (within the geomagnetic meridional plane, perpendicular to the field line, positive outward) between the storm day and quiet day is given in Figure 9. The satellite trajectories are the same as those in Figure 8. At 10:40 UT, the meridional ion drifts on the quiet day were slightly negative and inward between −80°W and −25°W (local time 05:12–09:15). However, the ionosphere on the storm day exhibited strong outward ion drifts between −40°W and −20°W (local time 08:05–09:31). At 12:20 UT, the meridional ion drift on the storm day between −55°W and −40°W (local time 08:40–09:50) was also positive and much larger than that on the quiet day and coincided with the horizontal ion drift in Figure 8.

From ~08:45 UT on 4 November, the high-latitude electric field increased significantly and penetrated into low latitudes (Figure 7). Then, the prominent outward E×B drift and southward horizontal ion drift appeared (Figure 8 and Figure 9), which uplifted the peak height of the ionosphere (Figure 6). Due to the low recombination rate at higher altitudes, the plasma accumulated and led to the positive ionospheric storm. If the HmF2 had been uplifted by the equatorward neutral wind, the horizontal ion drift would have been equatorward, contrary to the ICON observations in Figure 8. Thus, we conclude that the positive ionospheric storm was dominated by the electric field and thermospheric composition variations also contributed (Figure 5).

5. Conclusions

A positive ionospheric storm occurred from ~11:00 UT to ~18:00 UT on 4 November 2021 over the South American sector. Based on multi-instrument observations from GNSS, DMSP, GOLD, ionosondes, SuperDARN, and ICON, the driver of this positive storm was revealed. The main findings are summarized below:

1. The positive storm covered most of the South American sector, and the maximum TEC enhancement magnitude was about 100%.

2. In the southern hemisphere, significant ∑O/N₂ enhancement and depletion were observed at low and middle-to-high latitudes, respectively.

3. Compared with the quiet day, HmF2 during the positive ionospheric storm increased by up to 80 km, and an NmF2 increase appeared 3 h later.

4. Significant enhancement of the southern hemisphere CPCP was observed about one hour earlier than the HmF2 uplift.

5. According to the poleward and outward ion drifts, the HmF2 uplift was caused by PPEFs. The electric field effect dominated the positive ionospheric storm, and the thermospheric composition variation may have also contributed.