The SunRISE ground radio lab campaign: integrating student research, education, and outreach during the 2024 total solar eclipse

1 Introduction

The Sun Radio Interferometer Space Experiment (SunRISE) is a NASA mission of six CubeSats that will form a space-based low-frequency radio interferometer to image solar radio bursts and study particle acceleration associated with coronal mass ejections and solar energetic particles (Kasper et al., 2021). To complement this mission, the University of Michigan established the SunRISE Ground Radio Lab (GRL) ¹ (Figure 1; Akhavan-Tafti and Soni, 2025). The GRL develops ground-based radio instruments while providing a platform for undergraduate and graduate students to train in heliophysics research. Educational outcomes of the SunRISE GRL program, including team structure, pedagogy, mentoring framework, and student skill development, have been formally described and assessed in dedicated education-focused publications (Akhavan-Tafti et al., 2025; Akhavan-Tafti and Soni, 2025). This program is also collaborating with NASA’s RadioJove citizen science project ² (Fung et al., 2026). These works present the program’s educational design, including team-based learning, structured mentoring, and authentic research engagement, and provide evidence of gains in technical competency, teamwork, and problem-solving skills based on participants’ self-reports and program evaluation.

At the core of GRL’s activities are the Long Wavelength Array (LWA) antennas, portable systems operating in 40–160 MHz frequency band. These instruments are capable of detecting solar radio bursts, ionospheric variability, and background emissions, bridging the goals of the space mission with accessible, hands-on training opportunities.

Solar radio emission originates in the solar corona, where energetic electrons interact with magnetized plasma and generate radio waves at frequencies related to the local plasma frequency and its harmonics (Wild and McCready, 1950; Melrose, 1985). These emissions are generally categorized into continuous and transient (bursty) components. Continuous radio emission is associated with relatively steady coronal conditions and arises through processes such as thermal bremsstrahlung and gyroresonance, producing a slowly evolving radio background (Dulk, 1985). In contrast, solar radio bursts are short-lived, structured phenomena triggered by rapid energy release events, including magnetic reconnection, solar flares, and coronal mass ejections. Such bursts originate when energetic electron populations generate plasma oscillations that are subsequently converted into escaping electromagnetic waves, often appearing as drifting or fine-scale features in radio dynamic spectra (Reid and Ratcliffe, 2014).

Before reaching ground-based instruments, solar radio waves must propagate through the Earth’s ionosphere, which strongly influences low-frequency signals through absorption, dispersion, and refraction. Changes in ionospheric electron density can modify radio propagation, reducing signal strength or preventing transmission below the ionospheric cutoff frequency (Davies, 1990). During a solar eclipse, the temporary reduction in solar extreme ultraviolet (EUV) radiation decreases ionospheric ionization, leading to measurable changes in radio wave propagation and observed signal intensity (Coster et al., 2017). As a result, ground-based radio observations during eclipse conditions capture both the intrinsic characteristics of solar radio emission and the ionospheric response, providing a useful framework for investigating solar–terrestrial interactions.

Solar eclipses have long been used as natural experiments to study ionospheric response to rapid changes in solar radiation. Early radio studies showed that eclipses produce measurable reductions in ionospheric electron density, altering low-frequency radio propagation (Davies, 1990). More recent work using Global Navigation Satellite System (GNSS) observations has quantified eclipse-driven total electron content (TEC) depressions of several TEC unit (TECU), with the strongest effects occurring near the path of totality and delayed recovery times governed by recombination processes (Coster et al., 2017; Goncharenko et al., 2018). Low-frequency radio experiments have similarly reported reductions in received power during eclipses, attributed to changes in ionospheric absorption and refraction (Basu et al., 2010). In parallel, studies of solar radio bursts have established how coronal radio emissions are modified by ionospheric propagation effects at meter and decameter wavelengths (Dulk, 1985; Reid and Ratcliffe, 2014).

Although eclipse-related ionospheric disturbances have been extensively investigated using GNSS TEC, ionosondes, SuperDARN, and VLF propagation measurements (Pal et al., 2012; Biswas et al., 2023; Ghosh et al., 2023), fewer studies have combined ground-based VHF observations with independent ionospheric diagnostics. The present campaign combines dual-site SunRISE GRL LWA observations with GNSS-derived TEC measurements while documenting student participation in instrument deployment, data acquisition, and analysis. The scientific observations provide the context for examining how authentic field research can be integrated with heliophysics education and outreach.

The present study builds on this body of work by combining LWA radio observations with GNSS-TEC measurements during the 2024 solar eclipse, while emphasizing student-led data collection and educational outreach. The 8 April 2024, total solar eclipse presented an exceptional opportunity to deploy the GRL LWA antennas in a real-world campaign.

1.1 Campaign design and educational objectives

This paper presents the 2024 eclipse campaign as a case study that integrated research and education. The campaign was designed with two intertwined goals.

• Scientific Investigation: Record low-frequency radio variability during total and partial eclipse conditions and compare the observations with GNSS-derived TEC measurements.
• Educational Engagement: Provide hands-on training for undergraduate/graduate students (from different disciplines such as electronics, mechanics, and computer science) from the University of Michigan, Ann Arbor, MI, United States, in the Multidisciplinary Design Program (MDP), immersing them in the full cycle of an experimental heliophysics project. ³

Two deployment sites were chosen.

• Peach Mountain Observatory, Michigan (latitude: 42.4028° and longitude: 83.9247°), within the zone of partial eclipse with ∼95% obscuration, to examine weaker obscuration effects.
• Upland, Indiana (latitude: 40.4756° and longitude: 85.4944), along the path of solar eclipse totality, to observe radio and TEC variability along the path of totality.

At both observing sites, students were actively involved in transporting, assembling, and calibrating the SunRISE GRL LWA systems. They participated in real-time monitoring of data acquisition and later contributed to post-processing activities, including noise mitigation, intensity estimation, and comparisons with ionospheric variability. The campaign was intentionally structured to emphasize experiential learning by exposing students to the full workflow of a field-based scientific investigation, closely reflecting the environment of professional heliophysics research. In this study, we present observations from the dual-site deployment at Peach Mountain, Michigan, which experienced a partial eclipse, and Upland, Indiana, located along the path of totality. The dual-site deployment enabled comparison between total and partial eclipse conditions, allowing us to investigate how eclipse magnitude influences radio observations and ionospheric response. We discuss both the scientific observations and the educational value of student participation, illustrating how the campaign served as both a research effort and a hands-on learning experience.

2 Campaign overview

The 15-member Multidisciplinary Design Program (MDP) team from the University of Michigan, Ann Arbor, MI, was organized into specialized subgroups to maximize both technical contributions and student learning. Four coordinated teams were established: software, responsible for data acquisition pipelines, real-time visualization, and post-processing tools; mechanical, focused on antenna design, structural support, and deployment logistics; electrical, tasked with receiver electronics, power systems, and signal integrity; and system integration, which coordinated activities across all teams to ensure seamless operation of hardware, software, and field deployment. This collaborative structure reflected the organization of professional research environments, enabling students to develop expertise within their respective areas while contributing to the broader scientific and operational goals of the eclipse campaign.

The observing campaign was organized at two sites. The first was Peach Mountain Observatory, Michigan, which experienced a partial eclipse with ∼95% obscuration. The second was Upland, Indiana, located along the path of totality with ∼4 min of complete obscuration.

At both locations, students from the MDP led the deployment, calibration, and monitoring of the LWA antennas. Faculty and staff provided oversight, but data collection, troubleshooting, and real-time tracking were primarily student-driven. This structure created an authentic field-research environment that simulated professional scientific campaigns. As shown in Figure 2, Heliophysics MDP students actively participated in erecting the SunRISE GRL LWA antenna at Taylor University in Upland, Indiana, positioning it along the path of totality for the 2024 solar eclipse campaign.

3 Instrumentation: SunRISE GRL LWA antennas

The SunRISE Ground Radio Lab utilizes the LWA antenna design (Figure 3), a low-frequency instrument optimized for the 40–160 MHz range (Ellingson et al., 2009). This bandwidth is well-suited for detecting solar radio bursts (Types II, III, and IV) and monitoring ionospheric variability. For this campaign, the systems were configured for continuous wideband observations with a 1-s temporal cadence.

3.1 Instrument setup and signal chain

The primary observing instrument deployed during the campaign was a dual-polarization active Long Wavelength Array (LWA) antenna. The antenna consists of two orthogonal bow-tie dipoles, each composed of triangular aluminum frame elements approximately 1.5 m in length and 0.8 m in width, mounted at a 45° downward slope on a central mast. Structural support is provided through horizontal fiberglass braces connected to a steel center post. To maintain a stable beam pattern and reduce variations in ground impedance, the antenna was installed over a 3 m × 3 m galvanized steel wire ground mesh, secured flush to the surface using plastic stakes. The dipoles were geographically aligned, with one oriented in the north-south direction and the other east-west, enabling measurements sensitive to polarization-dependent radio features. Because the LWA antenna elements are fixed and exhibit a broad beam pattern, particularly at lower frequencies, observations relied on the solar transit through the stationary field of view rather than mechanical tracking.

Active impedance matching is provided by the LWA Front-End Electronics (FEE) housed at the apex of the mast. The FEE provides approximately 30 dB of gain, ensuring the system remains sky-noise dominated across the operating band. Radio Frequency (RF) signals from the FEE are transmitted via LMR-400 coaxial cabling to a weather-resistant, temperature-controlled equipment enclosure. The signal chain utilizes a Long Wavelength Array Power Coupler (LWAPC) to separate the RF signal from the DC power required by the active antenna. To bridge the frequency range of the LWA antenna (40–160 MHz) with the input range of the spectrometer, the signals are routed through a heterodyne up-converter. The frequency-shifted output is then digitized by a Callisto-NA spectrometer, a programmable heterodyne receiver designed specifically for solar radio monitoring. Data acquisition is managed via the e-Callisto software suite running on a dedicated data acquisition computer, which records dynamic spectra in Flexible Image Transport System (FITS) format. The system was configured to sweep the relevant frequency bands with a temporal resolution of 1 s, sufficient to capture rapid solar radio bursts.

3.2 Data collection and methodology

Data acquisition began several hours before first contact and continued beyond the end of the eclipse. At both sites, dynamic spectra were collected across the 40–160 MHz frequency range at 1 s cadence and stored locally for post-processing. The data are publicly available through the SunRISE GRL repository at SunRISE data archive. ⁴ Calibration included bandpass correction and removal of instrumental offsets. For analysis, the spectra were divided into three intervals: pre-eclipse, eclipse (partial/total), and post-eclipse, as well as incorporating GNSS-derived vertical TEC data for complementary comparison. Students participated in all phases of data acquisition, including antenna orientation, calibration, real-time monitoring, and post-processing, thereby gaining hands-on training in experimental heliophysics.

In addition to radio observations, ionospheric TEC was analyzed to investigate eclipse-driven variability. TEC values were obtained from GNSS measurements expressed in TEC units (TECU), where 1 TECU corresponds to 10¹⁶ electrons per square meter (Seemala, 2023; Radicella and Migoya-Orue, 2021). The dataset was sampled at a cadence of 15 min and examined over a five-day interval surrounding the eclipse, enabling comparison between the 8 April 2024, event and adjacent quiet days. This approach provided a day’s time frame reference for evaluating eclipse-induced departures from the normal diurnal TEC pattern at Indiana (total eclipse).

The LWA antennas are capable of detecting solar radio bursts, ionospheric variability, and background emissions (Akhavan-Tafti et al., 2025). Eclipse-related changes in solar illumination may alter ionospheric ionization and VHF propagation; these possible effects are evaluated in the Results and Discussion.

Persistent narrowband RFI frequencies were identified through visual inspection of the dynamic spectra and excluded from the integrated intensity calculation. Only relatively interference-free sub-bands were included to ensure that the resulting light curves primarily reflected broadband temporal variations rather than localized anthropogenic signals.

For this campaign, radio dynamic spectra were collected using the SunRISE GRL LWA antennas during the solar eclipse. To calibrate and analyze the observations, the dynamic spectra were processed by extracting intensity values from selected frequency sub-bands (Soni et al., 2022). Individual intensity profiles were generated for each sub-band, and a summed intensity curve was computed to provide a representative measure of overall radio variability at each observing site. This approach enabled both frequency-dependent and integrated analyses of radio signal variations during the eclipse interval.

For the purposes of this analysis, the observational timeline is categorized by the optical phases of the eclipse (partial and total). However, it is important to note that the solar corona, which is the source of the thermal radio emission and the EUV/X-ray flux responsible for ionizing the Earth’s upper atmosphere, extends significantly beyond the solar photosphere. Consequently, while the optical disk was 100% obscured during totality, the ionizing flux was not completely extinguished, and the ionospheric TEC response is expected to exhibit a complex temporal lag relative to the optical obscuration. Therefore, references to ‘totality’ in this text strictly denote the occlusion of the solar photosphere, rather than the cessation of all solar influence on the local radio environment.

3.3 Physical interpretation framework

The radio intensity recorded by the SunRISE GRL LWA systems during a solar eclipse is governed by several physical processes rather than eclipse obscuration alone. The received signal consists of (i) intrinsic radio emission originating from the visible solar disk, (ii) emission from the extended solar corona, which remains only partially occulted throughout much of the eclipse, and (iii) propagation effects introduced by the Earth’s ionosphere. As the Moon progressively occults the Sun, the relative contributions from the solar disk and the extended corona evolve continuously. At the same time, the reduction in solar extreme ultraviolet (EUV) radiation decreases ionospheric ionization, altering radio-wave propagation through changes in absorption and refraction, particularly within the lower ionosphere.

Consequently, the observed radio intensity cannot be interpreted solely as a function of eclipse obscuration. Instead, it reflects the combined influence of evolving solar emission geometry and ionospheric propagation. Throughout this paper, the LWA observations are therefore interpreted together with GNSS-derived total electron content (TEC) measurements to distinguish between source-related and ionospheric contributions to the observed variability.

Equation 1 provides a conceptual representation of the principal contributions to the measured intensity and is not used for quantitative fitting.where S(t) represents intrinsic radio emission from the visible solar disk, C(t) the contribution from the extended solar corona, P(t) ionospheric propagation effects (including absorption and refraction), T_ion (t) represents time-dependent ionospheric transmission, including absorption and refraction; and B residual background and instrumental or anthropogenic radio-frequency interference.

I obs t = S t + C t + T ion t + B , (1)

To facilitate interpretation of the observed radio intensity variations, Figure 4 presents a conceptual framework illustrating the primary processes contributing to the LWA measurements before, during, and after the eclipse. The received radio intensity is influenced by the combined effects of solar-disk emission, extended coronal emission, eclipse-induced ionospheric propagation, and background radio-frequency interference. During the eclipse, progressive occultation reduces the contribution from the solar disk, while a substantial fraction of the extended corona remains visible. At the same time, the reduction in solar EUV radiation modifies ionospheric electron density, altering radio-wave propagation. Consequently, the observed intensity should be interpreted as the combined response of these processes rather than as a direct measure of eclipse obscuration alone.

4 Results

4.1 Upland, Indiana (total eclipse)

Figure 5 presents the radio observations from Upland, Indiana. The top panel shows the dynamic spectrum recorded by the SunRISE GRL LWA antenna over the frequency range of 40–160 MHz. While no distinct spectral signatures can be directly attributed to the eclipse, the dynamic spectrum provides important observational context. Vertical dashed lines indicate the onset of the eclipse, the interval of totality, and the end of the event, establishing a temporal framework for interpreting variations in the integrated radio intensity.

The dynamic spectrum from Upland, Indiana, shows a persistent high-intensity band near 160 MHz throughout the observing interval before, during, and after the eclipse, with little change in intensity or structure. This stability suggests that the signal is not directly related to the solar eclipse, but is more likely associated with terrestrial radio-frequency interference (RFI) or a stable background emission within the instrument’s passband, such as local broadcast transmissions or instrumental band-edge effects.

In contrast, the short-lived enhancements near ∼120 MHz appear as isolated, burst-like features that are sporadic in time and confined to a narrow frequency range. These signals do not exhibit systematic timing with respect to the eclipse phases and lack the broadband or frequency-drifting characteristics typically associated with solar radio bursts. Consequently, they are more likely associated with transient radio-frequency interference (RFI), potentially originating from aircraft communications, radar systems, or other intermittent anthropogenic sources.

GOES soft X-ray measurements were examined over the period of eclipse. No flare was recorded during this interval, making flare-related X-ray enhancement unlikely to dominate the observed radio variability, suggesting that the observed variations were not driven by transient solar activity.

The bottom panel shows the corresponding integrated intensity curve. To generate the light curves, intensity values were extracted from selected frequency sub-bands, chosen to minimize contamination from persistent radio-frequency interference (RFI), and then summed to produce a representative intensity profile. This integrated curve provides a simplified view of temporal variations in the observed radio signal during the eclipse interval. Both the dynamic spectra and integrated intensity curves were generated at a high temporal resolution of 1 s cadence. Intensity variations are presented in relative units (e.g., relative power or voltage measured by the LWA system), with the pre-eclipse level used as the reference baseline.

The integrated intensity decreases rapidly during the early part of the observing interval and reaches its minimum before eclipse onset. It then remains comparatively low before increasing gradually through the eclipse interval and toward totality. The intensity reached a minimum nearly 20 units below the pre-eclipse baseline before slowly recovering. In terms of Equation 1, the observed profile may contain contributions from source geometry, ionospheric transmission, and residual background; their individual contributions cannot be quantitatively separated using the present measurements. No evidence for significant variations in the intrinsic solar emission term S(t) was found during the observing interval because GOES observations indicate an absence of flare activity.

Similar pre-eclipse variations have been reported in recent GNSS studies, which attributed them partly to background ionospheric evolution associated with local solar time rather than eclipse obscuration alone. The observations include eclipse onset and totality, but data acquisition ended before the complete post-eclipse recovery could be recorded because of site restrictions. However, observations could not be extended through the entire recovery phase because of site restrictions, which limited data collection beyond the total eclipse interval.

Unlike white-light eclipse observations, the measured radio intensity is influenced by both the visible solar disk and the extended corona. Although the lunar disk progressively obscures the photosphere, a substantial fraction of the coronal radio emission remains visible throughout the eclipse. Consequently, changes in the integrated radio intensity represent the combined effects of evolving solar source geometry and radio-wave propagation through the ionosphere rather than eclipse obscuration alone.

4.2 Peach Mountain, Michigan (partial eclipse: ∼95% obscuration)

Figure 6 presents the radio observations from the Peach Mountain Observatory, Michigan, where the eclipse was partial (∼95% obscuration). The top panel shows the dynamic spectrum from 40 to 160 MHz recorded by the LWA antenna. As with the Indiana observations, no clear eclipse-driven spectral features are visible, but the vertical dashed lines mark the timing of the eclipse start, maximum obscuration, and end, providing temporal context for interpreting the intensity variations.

The Peach Mountain dynamic spectrum shows a persistent narrowband signal just above 60 MHz that remains present throughout the observation period, suggesting a stable background or instrumental feature rather than an eclipse-related effect. In contrast, intermittent signals at higher frequencies appear sporadically and do not show a clear timing correspondence with the eclipse phases, making them more consistent with transient radio-frequency interference. These spectral features provide observational context, while any eclipse-related variability in the radio signal is more appropriately assessed through the integrated intensity measurements.

The bottom panel displays the summed intensity curve across multiple frequency bands. In contrast to Indiana, the Peach Mountain curve shows a smaller and noisier depression, on the order of ∼2–3 units relative to baseline. The decrease begins around the time of maximum obscuration, reaching its minimum shortly after 16:00 local time, and is followed by a gradual recovery. The subsequent recovery extended for nearly an hour beyond the third contact (end of solar eclipse), reflecting slower recombination processes in the ionosphere compared with the rapid depletion during totality. Superimposed on this overall trend are several short-term fluctuations, which may reflect local ionospheric variability or site-specific environmental effects.

Previous eclipse studies have similarly reported weaker and more spatially variable responses during partial eclipses owing to differences in eclipse magnitude, local time, and background ionospheric conditions (Gautam et al., 2024; Le et al., 2025).

Because the solar corona remains only partially occulted during the eclipse, its contribution to the received radio signal changes gradually throughout the observing interval. Within the framework of Equation 1, the weaker response observed at Peach Mountain is consistent with a smaller contribution from eclipse-induced propagation changes P(t), while continued visibility of the extended corona C(t) and local ionospheric variability contribute to the observed fluctuations.

4.3 GNSS-TEC observations

It is important to note that the GNSS TEC eclipse-related signal mainly comes from the ionospheric F-region, while LWA VHF signals are typically sensitive to the D/E regions or the integrated effect across the ionosphere. The GNSS-derived TEC measurements used in this study were obtained from global and regional GNSS receiver networks (Seemala, 2023), which provide continuous dual-frequency observations suitable for ionospheric analysis. In particular, TEC products were derived from data archived by the International GNSS Service (IGS) and associated regional networks. These datasets are widely used in heliophysics and space weather studies to quantify ionospheric variability on local to global scales, including responses to solar eclipses and geomagnetic disturbances (Radicella and Migoya-Orue, 2021, and references within).

GNSS-derived vertical TEC was analyzed near Upland, Indiana. Figure 7 shows the diurnal TEC variation for five consecutive days (April 6–10, 2024). On non-eclipse days (April 6, 7, 9, and 10), TEC followed a smooth diurnal pattern, increasing after sunrise, peaking near midday at ∼35–40 TECU, and declining toward nighttime values of ∼10–15 TECU.

On the eclipse day (April 8, red curve), the TEC exhibited a clear departure from this normal diurnal trend. During the eclipse interval, the April 8 VTEC profile departed from the surrounding-day diurnal behavior. Relative to the selected background reference, the eclipse-associated difference reached approximately 5–8 TECU. This decrease reflects the temporary reduction in solar EUV radiation and the associated decline in ionospheric electron production. While the GNSS TEC measurements clearly show an eclipse-related ionospheric response, the temporal variations in the LWA radio intensity curve do not exhibit a direct one-to-one correspondence with this TEC minimum.

This behavior reflects the sudden reduction in ionizing solar EUV radiation during eclipse totality, which reduced electron production rates in the ionosphere. The slower recovery is consistent with recombination processes dominating until solar input returned to full levels.

The apparent differences between the LWA intensity and GNSS-derived TEC measurements are expected because the two techniques probe different regions of the ionosphere. GNSS TEC primarily reflects electron density integrated through the F-region, whereas VHF radio propagation observed by the LWA is influenced by absorption and refraction within the lower D- and E-regions as well as integrated propagation effects. Consequently, the temporal evolution and magnitude of the two measurements need not correspond directly.

GOES X-ray observations show no significant flare activity during the observing interval, indicating that changes in radio intensity were unlikely to result from flare-associated EUV enhancement.

The delayed TEC minimum observed during the eclipse is consistent with previous studies demonstrating that the ionosphere does not respond instantaneously to reductions in solar radiation because recombination times in the F-region are significantly longer than photochemical response times in the D-region. Similar delays of 20–40 min have been reported by Gautam et al. (2024), Zhang et al. (2024), and Coster et al. (2017).

5 Educational and outreach impact

A total of 15 undergraduate and master’s students from the University of Michigan’s MDP played a central and active role in the solar eclipse campaign. The student team was organized into four coordinated subteams: software, electrical, mechanical, and system integration. This structure intentionally mirrors professional research environments, allowing students to develop domain-specific expertise while collaborating toward a shared scientific objective. The software team developed data acquisition and visualization pipelines, the electrical team managed receiver electronics and power systems, the mechanical team oversaw antenna design and field deployment logistics, and the system integration team coordinated communication and functionality across all technical domains.

Through this framework, students participated in every stage of the campaign. They were directly involved in transporting and installing antennas, calibrating instrumentation, monitoring real-time observations during the eclipse, and contributing to post-processing and preliminary data interpretation. This immersive participation extended learning beyond traditional classroom instruction, fostering technical competency, systems thinking, teamwork, and problem-solving skills under real field constraints.

In addition to radio-frequency monitoring, the student team conducted a parallel optical imaging campaign. This effort demonstrated their ability to manage multi-modal data acquisition in a time-sensitive field environment. The composite image sequence shown in Figure 8 includes visible solar prominences during totality, corresponding to active regions discussed in Section 4. The integration of optical and radio observations provided students with a broader understanding of solar activity and observational techniques.

The campaign also generated meaningful outreach engagement. Antenna deployments at Peach Mountain Observatory and Taylor University in Upland, Indiana, attracted interest from faculty, students, and members of the local community. Students played an active role in communicating the project’s goals, discussing solar radio science, ionospheric responses to eclipse conditions, and instrumentation design. Their visible participation in field operations highlighted the role of university-based research initiatives in heliophysics and provided opportunities to foster broader public interest in solar and space science.

At the time of manuscript preparation, 10 out of 15 participants were enrolled in graduate programs or had progressed to graduate study in engineering, physics, or computer science, and several continue to engage in heliophysics-related research projects. In the 18 months following the eclipse, students delivered several presentations based on the campaign data at venues including the Solar Heliospheric and INterplanetary Environment (SHINE) workshop 2025 in Juneau, Alaska, MDP Design Symposium, and departmental symposia. While the project did not include a formal pre- and post-assessment study, continued academic progression, research participation, and public dissemination activities provide measurable indicators of sustained engagement in STEM pathways. Informal post-campaign reflections collected from participants indicated increased confidence in conducting experimental research and a stronger interest in pursuing advanced study in space science. Evidence of the program’s educational impact is documented through student participation and outcomes reported on the official SunRISE GRL news platform. ^{5, 6, 7}

Among the student participants, several have continued into advanced academic and professional roles, reflecting sustained engagement in science and engineering. Leadership within the student team further illustrates the program’s educational value. During the eclipse campaign, Leo Greenleaf, then an undergraduate student in Mechanical Engineering and a member of the MDP System Integration Team, played a central role in coordinating field operations and contributing to data analysis efforts. As lead of the System Integration Team, he supervised the electrical, mechanical, and computer science subteams, ensuring coordination across technical domains and operational readiness at both observing sites. Leo currently serves as a research assistant supporting the GNSS expansion project, which aims to implement interferometric techniques to connect and integrate data collected from participating high schools. His continued involvement highlights the program’s role in fostering sustained participation in student-driven heliophysics research and instrumentation development.

Working within this integrated structure, Nolan Tribu, a member of the MDP System Integration Team and currently an undergraduate student in Space Sciences and Engineering at the University of Michigan, Ann Arbor, MI, United States, played a key operational role during the eclipse campaign. He coordinated cross-disciplinary efforts among the software, electrical, and mechanical teams to support antenna deployment and data acquisition at both observing sites. During field operations, he contributed to troubleshooting, real-time monitoring, and maintaining system stability during totality. Nolan’s leadership was later recognized through distinguished academic awards, reflecting his academic achievement, commitment to research, and service within the College of Engineering community. His contributions to cross-disciplinary coordination and real-time system integration demonstrate how project-based technical training within the SunRISE GRL program can support continued professional and academic development.

In parallel, Zhuo-yuan Liu, then a Master’s student in Electrical and Computer Engineering at the University of Michigan and now a PhD student in Computer Science at Florida State University, contributed substantially to the scientific analysis of the campaign data. He processed spectrogram observations collected during the eclipse, analysed corresponding GNSS-derived TEC measurements, documented solar radio bursts detected by the LWA system, and analysed properties associated with solar activity events. His work extended to the development of an LWA interferometry array concept and machine learning approaches for solar radio burst classification. Participation in the SunRISE GRL project strengthened his expertise in signal processing and data analysis and contributed to his continued academic progression into doctoral research.

The eclipse campaign therefore advanced the dual mission of the SunRISE GRL program by integrating scientific investigation with student training and workforce development. Through structured team organization, active student participation in authentic research tasks, leadership opportunities, and public engagement, the campaign illustrates how field-based heliophysics initiatives can support both scientifically meaningful observations and experiential learning opportunities for students interested in space science.

6 Discussion and conclusion

The 8 April 2024, total solar eclipse provided an opportunity to investigate eclipse-related radio and ionospheric variability using complementary LWA and GNSS observations while simultaneously engaging students in an authentic field-based heliophysics campaign. The dual-site deployment enabled comparison between total eclipse conditions at Upland, Indiana, and partial eclipse conditions at Peach Mountain Observatory, Michigan, providing insight into how radio observations respond under different eclipse geometries and ionospheric conditions.

The LWA dynamic spectra recorded at both sites were dominated by persistent narrowband features and intermittent localized enhancements, with no clear evidence of broadband or frequency-drifting solar radio bursts during the observing interval. GOES soft X-ray observations likewise showed no significant flare activity, indicating that transient flare-related enhancements in solar EUV or X-ray radiation were unlikely to influence the measurements. Consequently, the observed temporal variability is interpreted primarily in terms of the combined effects of solar source geometry, ionospheric propagation, and background radio-frequency interference.

As introduced in the conceptual framework (Equation 1), the measured radio intensity represents the combined contribution of intrinsic solar-disk emission, extended coronal emission, ionospheric transmission, and residual background signals. During eclipse progression, the Moon progressively obscures the visible solar disk, while a substantial fraction of the extended corona remains visible because the radio-emitting corona extends beyond the optical solar limb. At the same time, the reduction in solar EUV radiation modifies ionospheric electron density, altering radio-wave absorption and propagation. Consequently, the observed radio intensity cannot be interpreted solely as a function of eclipse obscuration but instead reflects the combined influence of evolving solar emission geometry and ionospheric propagation. As illustrated conceptually in Figure 4, the measured radio intensity reflects multiple simultaneous processes. This framework explains why the integrated intensity does not vary monotonically with eclipse obscuration and why differences are observed between the Indiana and Michigan sites despite their similar eclipse magnitudes.

The integrated intensity measured at Upland reached its minimum before the onset of the eclipse and increased gradually throughout the eclipse interval. This behavior demonstrates that the measured radio intensity does not directly track eclipse geometry. Within the framework of Equation 1, the observations suggest that temporal changes in ionospheric transmission, together with the evolving contribution of the extended corona and normal background variability, contribute significantly to the measured signal. Because no flare-associated solar radio activity was detected, variations in intrinsic solar emission are expected to be comparatively small during the observing interval.

The Peach Mountain observations exhibited smaller and more irregular variations than those recorded at Upland. The integrated intensity curve displayed multiple local minima and maxima rather than a monotonic decrease, indicating that local ionospheric conditions and propagation effects influenced the received signal. Similar spatial variability has been reported in previous eclipse investigations, where the ionospheric response depended not only on eclipse obscuration but also on latitude, local solar time, neutral atmospheric dynamics, and traveling ionospheric disturbances (Coster et al., 2017; Gautam et al., 2024; Zhang et al., 2024; Le et al., 2025). The differences between the two observing sites therefore illustrate the complexity of eclipse-induced ionospheric variability rather than simply the effect of eclipse magnitude.

The GNSS-derived TEC observations provide an independent perspective on the ionospheric response. Unlike the LWA observations, which are influenced primarily by radio-wave propagation through the lower ionosphere, GNSS-derived TEC represents the vertically integrated electron density and is dominated by the F-region ionosphere. Consequently, the two techniques probe different altitude regions and physical processes, and direct temporal agreement between their responses is not expected. The observed TEC reduction of approximately 5–8 TECU is consistent with the reduction in solar EUV-driven ionization during the eclipse and with previous GNSS studies that reported delayed TEC minima and extended recovery times associated with F-region recombination and transport processes (Coster et al., 2017; Gautam et al., 2024; Zhang et al., 2024). Rather than validating one another directly, the LWA and GNSS observations provide complementary diagnostics of eclipse-induced ionospheric variability across different regions of the ionosphere.

Previous eclipse investigations using VLF propagation measurements have demonstrated pronounced D-region electron density changes during eclipse conditions (Pal et al., 2012; Biswas et al., 2023; Ghosh et al., 2023). In contrast, the present study combines VHF radio observations with GNSS-derived TEC measurements, providing complementary sensitivity to both lower- and upper-ionospheric responses. This multi-instrument approach extends previous eclipse studies by illustrating how different observing techniques respond to distinct ionospheric regions and physical processes during the same eclipse event.

6.1 Limitations

Several limitations should be acknowledged. First, the present study does not quantitatively separate the individual contributions of solar-disk emission, coronal emission, ionospheric transmission, and residual radio-frequency interference to the measured LWA intensity. Equation 1 is intended only as a conceptual framework for interpreting the observations rather than as a quantitative model. Second, the Indiana observations concluded before the complete post-eclipse recovery was recorded because of site constraints. Third, the GNSS-derived TEC measurements primarily represent the F-region ionosphere and therefore cannot directly characterize the lower ionosphere that most strongly influences VHF radio propagation. Finally, observations were obtained from only two sites with different local environments and background conditions. Future work combining calibrated radio observations, forward modeling of solar radio emission, ionospheric ray-tracing, and additional diagnostics such as ionosondes or incoherent scatter radars will enable a more quantitative separation of source-related and propagation-related effects.

6.2 Educational and outreach outcomes

Beyond the scientific observations, the campaign demonstrated the value of integrating authentic research experiences into undergraduate and master’s education. Fifteen students from the University of Michigan Multidisciplinary Design Program participated in software, electrical, mechanical, and system integration teams, contributing to antenna construction, deployment, calibration, field operations, data acquisition, and post-processing. This organization mirrored the structure of professional scientific collaborations and provided students with practical experience in instrumentation, systems engineering, teamwork, and scientific problem-solving.

Several participants have continued into advanced academic or research roles following the campaign, illustrating sustained engagement with science and engineering. Students also presented campaign-related work at institutional and national scientific meetings, demonstrating continued involvement in research dissemination. Although no formal educational assessment was conducted, participant experiences and subsequent academic trajectories suggest that involvement in the campaign provided valuable opportunities for technical and professional development.

The campaign also generated meaningful public outreach. Antenna deployments at Taylor University and Peach Mountain Observatory attracted interest from students, faculty, and members of the local community, allowing participants to communicate the scientific objectives of the project and explain the principles of solar radio observations and eclipse-induced ionospheric variability. The visibility of these field activities highlighted the role of university-led research in advancing heliophysics while fostering broader public engagement with solar and space science.

7 Conclusion

This study presented the scientific and educational outcomes of the SunRISE Ground Radio Lab dual-site observing campaign conducted during the 8 April 2024, total solar eclipse. The LWA observations recorded complex, nonmonotonic radio-intensity variations at both observing sites that did not follow eclipse obscuration in a simple manner. The measurements are most consistently interpreted as the combined influence of evolving solar-disk and coronal emission, ionospheric transmission, background variability, and residual radio-frequency interference rather than eclipse geometry alone.

Complementary GNSS-derived TEC observations provided independent evidence of eclipse-associated F-region variability. Because the LWA and GNSS observations probe different ionospheric regions and physical processes, the two datasets should be regarded as complementary diagnostics rather than directly equivalent measurements. Together, they illustrate the importance of coordinated multi-instrument observations for investigating eclipse-driven solar-terrestrial interactions.

The campaign also demonstrated how a student-led field investigation can integrate scientific observations with education and outreach. Students participated directly in instrument deployment, calibration, field operations, data analysis, and scientific communication, gaining practical experience in experimental heliophysics while contributing to a coordinated observing campaign. Continued development of the SunRISE GRL network, including expanded interferometric capabilities and broader participation from university and high school partners, will provide additional opportunities to investigate solar radio phenomena while supporting workforce development and public engagement in heliophysics.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Footnotes

Edited by: Hyunju Kim Connor, National Aeronautics and Space Administration, United States

Reviewed by: Sujan Prasad Gautam, University of Delaware, United States

Sandip Kumar Chakrabarti, Indian Centre for Space Physics, India

Footnotes

URL: https://sunrise.umich.edu/,

https://radiojove.gsfc.nasa.gov/

https://sunrise.umich.edu/about-us/

https://sunrise.umich.edu/data-archive/(https://doi.org/10.7302/e4bh-mj50).

https://sunrise.umich.edu/news/,

https://news.engin.umich.edu/2025/06/us-high-schoolers-monitor-solar-storms-with-an-accessible-antenna-kit/

https://clasp.engin.umich.edu/2024/01/12/sunrise-ground-radio-lab-gives-high-school-students-the-skills-to-conduct-radio-science-and-astronomy/