2.1 Observations

To investigate ionospheric responses during the superstorm event over the equatorial region, Vertical Total Electron Content (VTEC) data from the Trivandrum Global Navigation Satellite System receiver part of the Indian Network for Space Weather Impact Monitoring were utilized (Choudhary et al., 2025). The VTEC values were derived using standard GNSS-based processing techniques as given in earlier studies (Choudhary et al., 2011; Rama Rao et al., 2006). In converting Slant TEC (STEC) to VTEC, only satellite signals with elevation angles exceeding $$50{}^{\circ}$$ were included. This constraint ensured that the Ionospheric Piercing Point at 350 km altitude remained within $$\pm 1.5{}^{\circ}$$ of the station's latitude and longitude. Vertical Total Electron Content was computed at 1-min intervals by averaging data from all visible PRNs that met the elevation threshold.

In addition to this, the electron density profiles and TEC (up to 1,000 km) were obtained using a DPS-4D digital ionosonde (Digisonde) located at Trivandrum which is an advanced digital ionospheric high frequency pulse sounding system developed at the University of Massachusetts Lowell's Center for Atmospheric Research (Reinisch & Huang, 2001). It sweeps through frequencies ranging from 1 to 18 MHz and provides reliable estimates of the electron density below the F-region peak using the “true height inversion algorithm” embedded in the SAO Explorer software package (http://umlcar.uml.edu/SAO-X/SAO-X.html). Digisonde-derived electron density profiles have been used extensively in the study of various solar-terrestrial phenomena (C.-C. Lee & Reinisch, 2007; Zong et al., 2010). In reality, ionosondes cannot measure densities above the ionization peak; however, a scale height approximation was used to infer the density of the plasma above the peak (Kutiev et al., 2009), and Ionospheric Electron Content (IEC) below 1,000 km is calculated.

2.2 Equatorial and Low-Latitude Physics-Based Stand-Alone Model

We compared the GPS (VTEC) and Digisonde (electron density profile and IEC) observations with simulations from a physics-based stand-alone model developed for equatorial and low-latitude ionospheric regions. The model solves the ion continuity and momentum equations along and across the geomagnetic field using an implicit finite difference scheme (Schunk & Nagy, 2009). Instead of solving the full energy balance equations, ion and electron temperature profiles are taken from external sources. This approach reduces computational cost but introduces uncertainties in the simulated electron densities. Such uncertainties are inherent to stand-alone models that do not self-consistently compute plasma temperatures, as variations in the input temperatures can directly affect the modeled electron density magnitudes.

The model incorporates photochemical reactions involving $${O}^{+}$$ (both ground, $${\mathrm{O}}^{+}$$(⁴$$S$$) and metastable states, $${\mathrm{O}}^{+}$$(²$$D$$) and $${\mathrm{O}}^{+}$$(²$$P$$)), $$N{O}^{+}$$, $${O}_{2}^{+}$$, $${N}_{2}^{+}$$, $${N}^{+}$$, $${H}^{+}$$, and $$H{e}^{+}$$ ions. It also accounts for perpendicular transport such vertical $$\vec{E}\times \vec{B}$$ drift of plasma and the parallel transport such as diffusion along geomagnetic field lines. Field-aligned geometry is incorporated via a dipole approximation, ensuring that transport processes align with magnetic field orientation. The current version of the model is restricted to the Indian equatorial–low latitude region by restricting the boundary conditions and provides electron density at every 1$${}^{\circ}\times 1{}^{\circ}$$ latitude-longitude bin and spans altitudes from 100 to 1,000 km.

For the present study, the model was run for two specific days: first, the geomagnetically quiet day 9 May 2024 $$(Ap=5)$$, and second, during the recovery phase of a geomagnetic storm, specifically on 11 May 2024 $$(Ap=271)$$, between 06:00 LT (00:52 UT) and 18:00 LT (12:52 UT). The quiet day model run is used as a reference to estimate the VTEC variations observed on the disturbed day as well as the efficiency of the model to reproduce the observations. The model-derived IEC was computed by integrating plasma densities from the model's lower boundary at 100 km up to 1,000 km.

Since the physics-based ionospheric model employed in this study operates as a stand-alone system, it requires external specification of key thermospheric and electrodynamic parameters. The neutral atmospheric background, including neutral densities and temperatures, is provided by the NRLMSIS 2.1 empirical model (Emmert et al., 2021, 2022). Inputs for ion composition, as well as ion and electron temperatures, are sourced from the International Reference Ionosphere (IRI-2020) model (Bilitza et al., 2022, 2024). Solar irradiance data are obtained from the Flare Irradiance Spectral Model version 2 (FISM2) (Chamberlin et al., 2008, 2020). Photoionization, photodissociative ionization, and photoabsorption cross sections, along with various chemical reaction rate coefficients, are taken from (Schunk & Nagy, 2009) and the publicly available database at http://phidrates.space.swri.edu. Among the most critical inputs to the model are the thermospheric meridional neutral wind and the equatorial $$\vec{E}\times \vec{B}$$ vertical plasma drift. These are discussed in detail in as follows:

Thermospheric meridional wind: Meridional winds significantly influence the equatorial and low-latitude ionosphere by modulating plasma diffusion along geomagnetic field lines. Poleward winds facilitate plasma transport from the magnetic equator toward low-latitude regions, enhancing diffusion along the field lines. In contrast, equatorward winds tend to suppress this diffusion, leading to distinctly different ionospheric responses (Balan et al., 2013; Fuller-Rowell et al., 1994; Rishbeth, 1971). The rate of change in the magnetic field aligned diffusion plays a critical role in determining the F-region peak height and the TEC above the equator (Ashok et al., 2024).

In this study, meridional wind data were obtained from the Thermosphere-Ionosphere Electrodynamics General Circulation Model (TIEGCM 2.0) for both quiet (9 May) and storm-time (11 May) conditions. We used TIEGCM outputs driven by the Assimilative Mapping of Ionospheric Electrodynamics (AMIE) technique, which enhances the realism of the model by assimilating high-latitude inputs. AMIE reconstructs storm-time electrodynamics by combining data from multiple sources–including ground-based magnetometers, SuperDARN radars, and satellite measurements from DMSP and NOAA–to estimate auroral precipitation and plasma convection patterns in the high-latitude ionosphere (Lu et al., 1998; Richmond & Kamide, 1988; Shiokawa et al., 2007).

Equatorial $$\vec{E}\times \vec{B}$$ vertical drift: Vertical plasma drift is an important electrodynamic parameter that governs the redistribution of ionospheric plasma. In this study, vertical drift values were derived using two distinct approaches corresponding to quiet and storm-time conditions. For quiet-time simulations, vertical drifts were obtained from the Scherliess–Fejer (SF) empirical climatological model (Scherliess & Fejer, 1999). This model, developed using observations from the Jicamarca Incoherent Scatter Radar and satellite-based Ion Drift Meter measurements, captures the diurnal, seasonal, and solar cycle variability of the equatorial vertical drift under low geomagnetic activity. Over the Indian sector, the variability in vertical drift can be significant, with standard deviations exceeding 5 m/s (B. G. Fejer, personal communication, 16 July 2024).

During storm-time conditions, vertical drifts were estimated using the PPEF model (Manoj et al., 2008, 2013), which accounts for rapid electric field variations driven by solar wind-magnetosphere interactions. This model utilizes a transfer function model to map data from the Interplanetary Electric Field (IEF$$y$$) to ionospheric electric fields. The transfer function was derived from a combination of IEF$$y$$ data from the ACE satellite, equatorial radar measurements from JULIA, and low-latitude magnetometer data from the CHAMP satellite. While the PPEF model performs well in reproducing daytime vertical drift enhancements over the Indian region, it tends to underestimate drifts during nighttime conditions (Manoj & Maus, 2012). Since the present study primarily focuses on daytime and post-noon VTEC behavior, this limitation is expected to have minimal influence on the results.

As the model relies on multiple empirical and physics-based inputs, care was taken to ensure internal consistency, particularly under disturbed conditions. Input parameters were incorporated at each time step and field-line position at fixed latitudes. Each parameter used a single input source, and its uncertainty propagated to the model output. Outputs were also cross-validated during quiet-time conditions to ensure continuity and compatibility. Sensitivity tests confirmed that the observed VTEC variations were not within the uncertainty limit of the input parameters.

4.1 Comparison of Simulated Electron Density Profiles With the Observation

Figure 2 presents a comparison of the temporal variation in the simulated electron density profiles for all four cases. As described earlier, Case 1 corresponds to a quiet-time scenario using the Scherliess–Fejer (SF) vertical drift and quiet-time meridional wind conditions (9 May). The results, shown in the top-left panel, exhibit smooth temporal variations, reflecting typical ionospheric behavior under undisturbed electrodynamic and neutral wind conditions.

To isolate the effect of PPEFs, Case 2 incorporates vertical drifts from the PPEF model while retaining the quiet-time meridional wind. The corresponding simulation, shown in the top-right panel, reveals an increase in both electron density and the height of the F-region peak. This confirms that storm-time PPEFs can drive substantial vertical plasma uplift. However, aside from this enhancement in density and peak height, no prominent structural undulations are observed in the electron density profiles when compared with Case 1. This indicates that the observed undulations in the electron density profiles cannot be attributed solely to PPEFs.

To assess the isolated influence of storm-time meridional winds, Case 3 employs the SF quiet-time drift combined with storm-time meridional winds obtained from TIEGCM, driven by high-latitude inputs from AMIE. The bottom-left panel shows the resulting electron density profiles. Notably, these simulations display undulations similar to those observed in the Digisonde measurements. While the comparison between Case 1 (SF drift + quiet-time wind) and Case 2 (PPEF drift + quiet-time wind) isolates the role of PPEFs, the comparison between Case 2 and Case 3 (SF drift + storm-time wind) highlights the influence of storm-time meridional winds in producing the observed undulations. This strongly suggests that the structural variations in the electron density are primarily driven by wind perturbations rather than electric field disturbances.

Case 4 represents the most realistic storm-time scenario, combining both PPEF-driven vertical drifts and storm-time meridional winds. The results, shown in the bottom-right panel, exhibit both enhanced electron density and the presence of undulations. When compared to Case 3, the main difference is the increase in electron density, confirming that PPEFs are the dominant driver of the storm-time density enhancements. However, the undulations arise only when disturbed meridional winds are included, underscoring their critical role in modulating the fine structure of the electron density profile.

4.2 Manifestation of the Electric Field and Neutral Wind Disturbances in VTEC

As explained in Section 3, the VTEC over Trivandrum showed unusual enhancement and prominent undulations on 11 May compared to the quiet day, 9 May. Model simulations indicate that electron density enhancements are reproduced when PPEF is incorporated, whereas the undulations emerge primarily when storm-time disturbed winds are included. To assess the impact of these disturbances on the TEC over the equator, the IEC was computed by vertically integrating the simulated electron density profiles in the altitude range of 100–1,000 km. The resulting IEC captures how different combinations of electric fields (vertical drifts) and thermospheric winds influence the ionospheric response during the geomagnetic storm. As before, four simulation cases were considered, and all scenarios are presented in Figure 3.

The teal line represents the IEC under quiet-time conditions (Case 1), showing a smooth temporal evolution. The orange line (Case 2) includes PPEF, where the IEC exhibits a modest increase, evident when compared with the teal line. With storm-time disturbed winds (Case 3), the IEC closely matches the GPS-derived VTEC, both in magnitude and in the presence of undulations until local noon, though the model underestimates VTEC in the afternoon. When both PPEF and storm-time winds are included (Case 4), the IEC shows a further increase, and the undulations are well reproduced. This confirms that PPEF mainly contributes to the overall IEC enhancement, while disturbed meridional winds predominantly drive the undulations.

In addition to GPS-derived VTEC and simulated IEC, Figure 3 also includes IEC values derived from Digisonde data (green line). There is a remarkable agreement between the model-simulated IEC and the Digisonde IEC, especially after 13:00 LT, during which the GPS VTEC shows substantial enhancement. Both the model and Digisonde integrate electron density between 100 and 1,000 km, whereas GPS VTEC is obtained along the slant path of the radio signal, corrected for elevation angle. Therefore, the net increase in GPS VTEC during the afternoon is likely associated with plasma content above 1,000 km–a component that is not captured by either the digisonde or the model.

Notably, on 11 May 2024, between 11:00 and 12:00 LT, ionograms from Trivandrum revealed F3 layer traces drifting above 1,000 km (Ambili et al., 2025; Ashok, Ambili, & Choudhary, 2025). Such features lie beyond the Digisonde's detection range and are not represented in the present modeling framework, contributing to the underestimation of TEC during this interval. Earlier studies have highlighted the role of plasmaspheric electron content in augmenting GPS TEC (Lunt et al., 1999; Yizengaw et al., 2008), showing that under storm-time conditions a substantial fraction of TEC can originate from the plasmasphere, particularly at equatorial and low latitudes, where the GPS raypath traverses a longer plasmaspheric segment than at mid- or high-latitudes. The mismatch observed after 13 LT thus underscores the need for extended vertical coverage in both modeling and observations to fully capture the ionosphere–plasmasphere system over the Indian dip equatorial region.

4.3 Variabilities Observed in the Meridional Wind on 11 May

In the previous section, we concluded that storm-time meridional winds are the primary driver of the observed undulations in electron density profiles and GPS-derived VTEC. This variability is illustrated in Figure 3. The bottom panel shows the latitudinal distribution of TIEGCM-simulated meridional winds at 350 km altitude between 06:00 and 17:00 LT, with red dots marking the location of the magnetic dip equator (Trivandrum). The top-middle panel presents the temporal evolution of the meridional wind over Trivandrum, while the top-right panel compares modeled IEC with GPS VTEC and Digisonde-derived IEC. Cyan-shaded intervals indicate times when the near-equatorial meridional wind reverses from northward to southward, a transition crucial for plasma redistribution.

The observed meridional wind undulations are likely linked to storm-time variability in geomagnetic activity. Geomagnetic disturbances enhance auroral particle precipitation and Joule heating, which increase high-latitude thermospheric pressure gradients and drive stronger equatorward winds (Lu et al., 2016; Richmond & Lu, 2000; Smith et al., 1982). Consequently, short-term variations in geomagnetic indices can modulate large-scale circulation patterns and produce the wind variability observed here.

Between 06:00 and 08:00 LT, the meridional wind is directed northward over Trivandrum and southern latitudes, and southward over northern latitudes. This convergence zone decelerates plasma diffusion along magnetic field lines, promoting plasma accumulation around the magnetic dip equator and leading to substantial density enhancement. Ambili et al. (2025) reported that VTEC increased by nearly 100% compared to the quiet day (9 May). Our analysis confirms that this early-morning enhancement in VTEC was driven by meridional wind convergence toward the dip equator.

During 08:00–09:00 LT, the meridional wind becomes strongly southward over Trivandrum. Correspondingly, a significant reduction in IEC is observed in the model simulations, accompanied by a delayed response in the Digisonde-derived IEC. However, this reduction is not evident in the GPS VTEC, although the standard deviation of VTEC during this interval is notably high. It is plausible that the dip in VTEC may have been present in individual PRNs (satellites) but was averaged out when data from multiple PRNs at different azimuths and elevations were combined.

As the day progresses, a prominent reversal in meridional wind direction begins around 10:00 LT, becoming predominantly southward between $${\sim} $$ 11:00 and 15:00 LT. These reversal periods (marked by cyan shading) coincide with significant undulations in the VTEC over the dip equator. The correspondence is evident in both modeled and observed data, suggesting that reversals in meridional wind direction directly influence vertical plasma transport, thereby modulating the TEC profile. The southward wind during this period likely pushes the plasma away from Trivandrum toward southern latitudes, resulting in noticeable dips in VTEC. These two pronounced VTEC undulations are closely synchronized with the timing of meridional wind reversals.

A notable feature in the analysis is the delayed response of Digisonde-derived IEC to the southward turning of the meridional wind between 10:30 and 12:00 LT (Figure 3, top-right panel). In contrast, the simulated IEC shows a more immediate decrease following the wind reversal. This discrepancy arises from observational limitations, such as the temporal resolution or sampling cadence of the Digisonde, and from uncertainties in model input parameters that affect the timing and amplitude of the simulated response. These factors together likely account for the lag observed in the Digisonde IEC relative to the model.

It is also noteworthy that during intervals of southward wind, a marked reduction in VTEC was reported over Bhopal (EIA crest location) (Ambili et al., 2025; Jain et al., 2025). This regional pattern underscores the broader influence of meridional wind dynamics on plasma redistribution across the Indian sector, effectively suppressing the formation of the EIA. Previous studies have shown that equatorward daytime winds (i.e., southward winds in the Northern Hemisphere) can suppress upward drifts and alter equatorial plasma distributions (Lin et al., 2005; Maruyama et al., 2005).

The one-to-one correspondence between meridional wind reversals and VTEC fluctuations provides strong evidence of a causal relationship, highlighting the role of meridional winds in short-term, localized modulation of the equatorial TEC. Thus, the pattern of wind reversal emerges as a key driver of the short-term variability in equatorial VTEC. The observed VTEC undulations from morning through afternoon are not incidental but are directly linked to dynamic changes in meridional winds. In contrast, the role of storm-induced electric fields appears to be limited to modulating the overall magnitude of VTEC, rather than its short-term structure.

Comparable daytime VTEC enhancements were reported during the 29–31 October 2003 Halloween storm, primarily attributed to PPEFs (Mannucci et al., 2005; Tsurutani et al., 2008). In contrast, the undulatory VTEC features observed during the 2024 storm closely associated with rapid meridional wind reversals are less frequently documented, particularly over the Indian sector. Although storm-time wind effects have been examined previously (Fuller-Rowell et al., 1994), clear observational evidence of wind-driven VTEC modulations at this resolution has been lacking. The present results thus provide the first detailed evidence of such variability, underscoring the critical role of meridional winds in shaping the low-latitude ionospheric response to geomagnetic storms.