info@apippress.com
+86 18163351462 (WhatsApp)
+86 182 8014 0209
Article Content
1. Introduction
The influence of solar eruptive and radiative processes on the near-Earth space environment has been investigated for decades to understand the mechanisms through which solar activity affects processes within the Earth’s atmosphere, thermosphere, and ionosphere [1,2]. Due to the increasing demand for near-Earth operations with low Earth orbit and middle Earth orbit satellites for monitoring, positioning, and navigation, the geomagnetic variability of the ionosphere and plasmasphere is the subject of utmost academic attention. Studies focusing on extreme solar events, such as flares and coronal mass ejections (CMEs), are particularly important. These events often lead to extended periods of heightened geomagnetic activity and energetic particle precipitation, causing significant changes in the thermosphere/ionosphere and impacting technologically advanced ground-based and space-borne systems [3,4]. Examples illustrate the potential severity: a powerful Earth-directed solar event can significantly disrupt terrestrial systems. Notable instances include the degradation of Global Positioning System (GPS) navigation performance during geomagnetically disturbed conditions [5,6]; the electric power blackout in Quebec, Canada, during the 13–15 March 1989 storm [7]; the electric power blackout in southern Sweden on 30 October 2003 [8]; and the loss of thirty-eight Starlink satellites due to a geomagnetic storm on 4 February 2022, resulting in significant financial and aerospace consequences [9]. These incidents underscore the need for continued study to better understand the characteristics and evolution of extreme space weather events.
The geomagnetic storm of 10–11 May 2024 has recently received considerable attention [10,11,12,13,14,15,16,17,18,19]. The storm impacted power grids, disrupted precision navigation systems, and generated widespread auroral displays. It significantly affected the composition, temperature, and dynamics of the Earth’s thermosphere, as observed by the Global-Scale Observations of the Limb and Disk (GOLD) instrument [11]. The GOLD revealed significant thermospheric heating, a decrease in the column-integrated O/N2 ratio (ΣO/N2) at high latitudes, and a complex morphology in temperature and ΣO/N2 at mid- and equatorial latitudes. The authors of [11] attributed the latter to potential longitudinal variations in meridional winds; however, a lack of direct wind measurements precluded confirmation of this hypothesis.
Several studies have examined the ionospheric response to the 10–11 May 2024 geomagnetic storm in various geographical regions [9,10,11,12,13,14,15,16,17]. For instance, analyzing global ionospheric total electron content (TEC) maps, Ram et al. [12] reported a substantial ionospheric disturbance, including a TEC increase exceeding 100% during the daytime on 10 May in the American longitudinal sector (75° W). They showed that this TEC enhancement occurred from low to mid-latitudes (±45°) in both hemispheres, resulting from the intensification and poleward expansion of the Equatorial Ionization Anomaly (EIA) crests. Similarly, Singh et al. [13], analyzing GPS TEC data over the American sector and incoherent scatter radar data from Jicamarca, Peru, observed an anomalous enhancement and expansion of the EIA to mid-latitudes (up to 38° N and 50° S). They reported large TEC increases during the afternoon and nighttime of 10 May at low and mid-latitudes (up to 1325% in the Southern Hemisphere and 380% in the Northern Hemisphere), persisting for approximately 8 h until 23:00 LT over the Americas.
In contrast, Kwak et al. [15] investigated the ionospheric response using ionosonde (foF2) and GNSS TEC data from stations in the East Asian region (Korea, Japan; 127.1° E–141.8° E, ~22.5° N–37.4° N dipole latitude). Their observations indicated a strong negative ionospheric storm effect, characterized by significant decreases and intense fluctuations in foF2 and TEC on 11 May. Jain et al. [16] also reported a dominant negative phase on 11 May using GPS TEC data from the Indian region (Bhopal, 14.2° N GMLAT), with a maximum deviation of approximately −68.5% around 20:45 UT. However, they also observed positive effects, including a +61% deviation around 05:00 UT on 12 May, TEC enhancements (24% to 50%) during the main phase, and quasi-periodic structures on 12 May, interpreted as signatures of prompt penetration electric fields and storm-induced traveling ionospheric disturbances (TIDs). Pierrard et al. [17], using GPS TEC data over Europe, observed an initial brief ionization increase followed by a rapid and prolonged TEC decrease from 10 to 12 May across northern (61° N), middle (50.5° N), and lower European latitudes (36° N).
These studies indicate that the 10–11 May 2024 extreme geomagnetic storm caused significant spatiotemporal ionospheric variations, exhibiting complex morphology dependent on geographic location. A comprehensive understanding of the global ionospheric disturbance necessitates analyzing data from diverse regions. This study aims to investigate the ionospheric response to this major geomagnetic storm (SYM-H minimum = −518 nT) using TEC data from the Central Asian (Chumish station) and East Asian (Sheshan, Fangshan, Chungchan, and Daejeon stations) regions.
Section 2 describes the data and methods. Section 3 presents the results and discussion. Section 4 provides the conclusions.
2. Materials and Methods
The ionospheric response to the 10–11 May 2024 geomagnetic storm was investigated using total electron content (TEC) data derived from Global Positioning System (GPS) receivers located at stations in the Central Asian (CAR) and East Asian (EAR) regions (Table 1). The geographic coordinates of the stations were obtained via the Nevada Geodetic Laboratory service (https://geodesy.unr.edu/NGLStationPages/GlobalStationList, accessed on 7 July 2025). The geomagnetic latitudes were calculated according to the geographic coordinates of the stations using the IRI-Plas model via the IONOLAB service (Department of Electrical and Electronics Engineering, Hacettepe University, Ankara, Türkiye; http://www.ionolab.org).
Table 1. GPS stations used in this study.
TEC is defined as the line integral of electron density along the path between the satellite and the receiver. It corresponds to the total number of electrons in a column of 1 m2 cross-section. The unit of TEC is TECu, and 1 TECu = 1016 el/m2. TEC has been widely recognized as a reliable indicator of ionospheric variability, and the cost-effective estimation of TEC using GPS data has been used in ionospheric studies since 1998.
The TEC values in this study were estimated using IONOLAB-TEC software (v1-41), which is available both online at www.ionolab.org and is also in an executable file form that can be downloaded to local computers. The online version requires only the date and the GPS receiver station name that can be chosen from the pop-up map. It automatically downloads the necessary RINEX, IONEX, and Ephemeris files, and the estimated TEC can be viewed and compared with the TEC estimates of IGS Ionospheric Analysis Centers. The manual for operating the executable file is provided along with a detailed description. If desired, the Slant TEC values can also be viewed in an executable file form on the user’s computer. The executable file also works automatically on the user’s computer, requiring only the date and file depository address. All other necessary files were downloaded in a user-friendly manner. The output files were written into user-defined directories. In this manner, the IONOLAB-TEC used in this study was generated by entering the station names and dates on the website or in the user-defined directories on their computer. The main algorithm of IONOLAB-TEC is the scientifically reproducible Reg-Est algorithm, as discussed in detail in [20]. If the user wants to write the code by themselves, then all of the equations are provided in the open literature. The Reg-Est is a robust method developed for estimating vertical total electron content (VTEC) from GPS measurements at a high temporal resolution of 30 s. Its function involves combining slant TEC data, derived from either pseudo-range or the less noisy phase-corrected measurements, gathered from all GPS satellites visible above a 10° horizon limit at a specific receiver location over a desired period [21]. The algorithm operates by minimizing a cost function through a least-squares approach; this function incorporates a high-pass penalty filter, enabling the accurate representation of sharp, sudden temporal variations in the ionosphere. Furthermore, Reg-Est can utilize optional weighting functions to mitigate multipath effects, particularly from low-elevation satellites, and appropriately handle differential code biases [22]. The algorithm uses IONOLAB-BIAS for receiver differential code bias estimation [23]. IONOLAB-TEC provides robust, high-resolution TEC values that capture important temporal features across diverse geomagnetic conditions and various geographic locations. Essentially, IONOLAB-TEC offers the detailed computational technique necessary for deriving high-fidelity TEC values from raw GPS data, likely serving as the foundation for services that provide accurate TEC products, such as those potentially offered by IONOLAB-TEC online to all researchers at www.ionolab.org, either as a downloadable executable file or online through a user-friendly interface [24]. The data used in this study were obtained via the IONOLAB service (Department of Electrical and Electronics Engineering, Hacettepe University, Ankara, Türkiye; http://www.ionolab.org), with a temporal resolution of 2.5 min. IONOLAB-TEC has been used in various studies, and it is one of the most cited contributions to this area. The ability of IONOLAB-TEC to estimate single station TEC with significant accuracy, reliability, and robustness has made it one of the most important software programs used in a wide variety of applications, including but not limited to modeling, mapping, and tomography [22,23,24].
The analysis period spanned 10–14 May 2024, with 9 May 2024 selected as a geomagnetically quiet reference day (Ap = 5, Kp = 1.3). Solar wind parameters and interplanetary magnetic field (IMF) data were used to characterize the drivers of the storm. Specifically, data on the total IMF strength (Bt), the north–south IMF component (IMF-Bz), solar wind speed (Vsw), and solar wind dynamic pressure (Pdyn) were obtained from the NASA/GSFC OMNIWeb database (https://omniweb.gsfc.nasa.gov/form/omni_min.html, accessed on 7 July 22025). Geomagnetic conditions were assessed using the planetary Ap and Kp indices, the Auroral Electrojet (AE) index, and the symmetric horizontal component (SYM-H) index, also sourced from the OMNIWeb database. The general space weather context for May 2024 was obtained from NOAA/SWPC weekly reports (ftp://ftp.ngdc.noaa.gov/STP/swpc_products/weekly_reports/, accessed on 7 July 2025) and SpaceWeather.com (http://spaceweather.com). In the next section, we provide detailed results and discussion.
3. Results and Discussion
3.1. Solar Wind Parameters and Geomagnetic Indices
Geomagnetic storms are disturbances in the Earth’s magnetosphere resulting from interactions with perturbed solar wind flows. Storm analysis typically distinguishes three phases: (1) the initial phase often marked by a sudden commencement (SC), or storm sudden commencement SSC, or sudden impulse (SI); (2) the main phase, characterized by a significant decrease in geomagnetic indices like Dst or SYM-H; and (3) the recovery phase, during which the indices gradually return toward pre-storm levels [25].
The initial phase of a geomagnetic storm is typically triggered by the arrival of an interplanetary shock wave or the leading edge of a CME impacting the magnetosphere, causing a compression often registered as an SC (if followed by a main phase) or an SI.
In May 2024, NOAA Active Region (AR) 3664, known as NOAA Active Region 13664 according to the NOAA/SWPC region number, was the primary source of solar activity, producing numerous M-class (37) and X-class (9) flares, along with associated CMEs, primarily between 8 and 12 May [11,26]. The X-class flares on 8–9 May likely initiated the CMEs responsible for the subsequent complex interaction with the magnetosphere and the resulting geomagnetic storm on 10–11 May 2024. The initial arrival was marked by a sudden impulse (SI) of 108 nT recorded by the Boulder magnetometer at 16:45 UT on 10 May.
Figure 1 presents UT variations in SYM-H, IMF Bt, IMF Bz, solar wind speed (Vsw), dynamic pressure (Pdyn), and the AE index for 10–12 May 2024. The SYM-H panel indicates an initial positive excursion, associated with the SC, beginning around 17:05 UT on 10 May and lasting approximately 48 min (peak SYM-H: 88 nT at 17:15 UT). The main phase’s development, characterized by a sharp decrease in SYM-H, commenced roughly three hours after the initial impulse. SYM-H then decreased rapidly over approximately 8 h, reaching a minimum value of −518 nT at 02:14 UT on 11 May (indicated by the solid vertical line in Figure 1). Subsequently, SYM-H began to increase, marking the onset of the recovery phase.
Figure 1. UT variations in SYM-H, IMF-Bt, IMF-Bz, solar wind speed (Vsw), dynamic pressure (Pdyn), and the AE index for 10–12 May 2024.
During the initial phase (SC), Figure 1 shows a surge in IMF-Bt intensity from ~5 nT to over 30 nT, coinciding with a southward turning of IMF-Bz. Concurrent increases in solar wind speed (from ~450 km/s to ~700 km/s), dynamic pressure, and AE activity were observed, likely associated with the arrival of multiple CMEs [26] carrying strong southward interplanetary magnetic fields. The main phase coincided with elevated IMF-Bt (up to approx. 70 nT), strong IMF-Bz fluctuations (±50 nT) predominantly southward, further increases in solar wind speed and dynamic pressure, and extremely high auroral activity (AE ≈ 3500 nT).
The recovery phase, as depicted in Figure 1, coincided with further intense AE activity, peaking around 09:00 UT on 11 May (AE up to 4000 nT), followed by a gradual decrease toward pre-storm levels by ~06:00 UT on 12 May. During this phase, IMF-Bt decreased, but IMF-Bz remained predominantly southward until ~18:00 UT on 11 May, with intermittent strong southward excursions (down to approx. −30 nT). Solar wind speed continued to increase, peaking near 1000 km/s around 02:00 UT on 12 May. Multiple decreases in SYM-H during the recovery phase coincided with intervals of negative IMF-Bz.
Overall, the analysis suggests the initial storm phase resulted from the increased dynamic pressure upon shock/CME arrival. The main phase developed under conditions of strong IMF magnitude, predominantly southward IMF-Bz, and high dynamic pressure. The total duration of intense interplanetary and geomagnetic activity (from SC until ~02:00 UT on 12 May) coincided with the period of increasing solar wind speed. The storm’s main phase development and duration were closely linked to the presence of strong southward IMF-Bz. Notably, the minimum SYM-H value occurred approximately 1.5 h after the minimum IMF Bz value (−46 nT at 00:41 UT on 11 May).
3.2. Observed Effects in the Ionosphere
A detailed discussion of the complex mechanisms driving ionospheric storm responses is beyond the scope of this paper; comprehensive reviews are available in [27,28,29,30,31,32]. However, a brief overview of the mechanism typically responsible for negative ionospheric storm phases, particularly relevant for the Northern Hemisphere in May (summer-like conditions), is warranted.
During geomagnetic disturbances, energy deposition (primarily Joule heating from currents and particle precipitation) in the auroral zones heats the lower thermosphere (100–140 km) [27,29,32]. This heating causes atmospheric upwelling, leading to changes in thermospheric composition at F-region altitudes (~150–600 km). Specifically, the ratio of atomic oxygen density [O] to molecular nitrogen [N2] and oxygen [O2] densities decreases (i.e., with increasing average molecular mass, the [O]/[N2] ratio is reduced) [27]. Since the ion production rate is proportional to [O], and the dominant loss rate for F-region electrons is proportional to [N2] and [O2], a reduced [O]/[N2] ratio leads to decreased electron density (Ne). Additionally, heating generates pressure gradients that drive a storm-induced equatorward circulation. This circulation transports thermospheric air with altered composition (low [O]/[N2]) to lower latitudes, causing the negative ionospheric phase (regions of depleted Ne) to extend equatorward [27,32]. Furthermore, the heated thermospheric gas with reduced [O]/[N2] also has an elevated temperature (T), which increases the chemical loss rate coefficients at F-region altitudes, further contributing to the Ne decrease [32,33]. Thus, negative storm phases are often attributed to the combined effects of reduced [O]/[N2] and increased T [34]. In the summer hemisphere, the prevailing background thermospheric circulation is generally equatorward, which reinforces the storm-induced circulation, facilitating the transport of compositionally disturbed air to mid- and low latitudes [27,32]. Therefore, in the summer hemisphere, negative storm effects can manifest during both daytime and nighttime. Impulsive heating events can initiate traveling atmospheric disturbances (TADs), which manifest in the ionosphere as traveling ionospheric disturbances (TIDs) [28].
Figure 2 shows the TEC variations observed at the Chumish station (CAR region), along with the AE index (Figure 2a) representing geomagnetic activity. The thick solid line shows the observations, while the thin line represents the TEC on the quiet reference day (9 May) for comparison (Figure 2b). The vertical dashed line indicates the SC time. Shortly after the SC (~18:00 UT or 23:00 LT on 10 May, approx. 1 h post-SC), a significant TEC decrease occurred, leading to a prolonged period of depleted TEC, characteristic of a negative ionospheric storm on 11 May (Figure 2b–d). This contrasts with observations in the American sector under similar daytime conditions, where the storm onset featured a sharp TEC increase [12,13]. The deviation from the quiet day (ΔTEC = TEC disturbed − TEC quiet) at Chumish reached a maximum depletion of approximately −37 TECu (Figure 2c) around 02:14 UT (07:14 LT) on 11 May, corresponding to a percentage decrease of about 70% (Figure 2d). This maximum depletion occurred approximately 9 h after the SC and coincided with the minimum SYM-H value. The duration of the most intense part of this negative phase seemed to agree with the duration of southward IMF Bz activity (cf. Figure 1, Bz panel). The total duration of the negative phase, including the recovery, exceeded three days and generally followed the trend of elevated AE activity (Figure 2a). This suggests that the altered thermospheric composition induced by the storm persisted throughout this period. Fluctuations are evident throughout the negative phase (Figure 2c,d), indicating the presence of large-scale irregularities and a disruption of the typical ionospheric structure during both the main and recovery phases. Overall, the ionospheric response observed in Chumysh generally corresponded to the above-described scenario of the development of the negative phase of an ionospheric storm in the summer season, caused by corresponding changes in the composition and circulation of the thermosphere in the middle latitudes.
