Anna Milillo

During BepiColombo second Mercury’s flyby on 23 June 2022, a ground-based campaign with the THEMIS solar telescope provided simultaneous observations of the sodium (Na) exosphere while Bepicolombo measured interplanetary and magnetospheric conditions. Between 8:53 and 9:34 UT, the exospheric Na emission abruptly increased in intensity by about 70% across the entire day side, followed by a slow recovery to the preceding level over approximately four hours. BepiColombo did not register neither relevant perturbations of the solar wind nor solar energetic particles during the Na increase. The configuration of Mercury, Earth and Solar Orbiter enabled observations of the solar regions facing Mercury from both sides. They revealed that an active region facing Mercury, but hidden from Earth, produced a flare and an extreme-ultraviolet (EUV) brightening when the Na emission increased, whereas subsequent emissions from post-flare loops produced no further increase. While photon-stimulated desorption is one of the main mechanisms releasing Na from the surface, short-therm Na variability has so far been attributed mainly to plasma precipitation. Our observations suggest that solar flares can trigger rapid Na variability at Mercury, and point to a multi-phase release from a finite reservoir of surface sodium, reminiscent of the steeper than linear dependence on solar EUV reported for the Moon on longer timescales. This result marks a significant step forward in understanding the mechanisms governing the interaction of Mercury and the Moon with the Sun and it was possible thanks to multi-messenger observations highlighting the potential of investigations during BepiColombo nominal mission.

Mercury is the smallest planet in the solar system with a radius of RM=2440 km and a large conductive core of approximate radius RC=0.8 RM [1, 2]. Despite its small stature, Mercury possesses a dynamo generating a dipole-like field with a magnetic moment of 195±10 nT-RM3 that is offset ~0.2 RM north [3, 4]. The combination of variable solar wind dynamic pressure due to Mercury's eccentric orbit and the planet’s occupancy of a large fraction of its asymmetric, weak magnetosphere produces a system in which magnetosphere topology and particle precipitation onto the surface are sensitive to both external driving and internal magnetic structure [7, 8, 9, 10]. The Hermean magnetosphere is also influenced by electromagnetic induction due to the presence of Mercury’s electrically conducting core, which modifies the magnetic response due to time-variable external forcing [10, 11, 12]. Previous work has shown that induced currents can strengthen the effective field during solar-wind pressure enhancements, altering magnetopause stand-off distance and compression of the dayside magnetosphere [11, 13, 14, 15, 16]. We use the Amitis GPU-based hybrid code [17, 18] to investigate the influence of a conductive core on magnetosphere topology and surface precipitation during an extreme solar event under different interplanetary magnetic field (IMF) orientations and interior configurations. We find that across all simulated cases, IMF orientation is the dominant first-order control on where reconnection can occur and on the large-scale reorganization of precipitation patterns, which agrees with previous work [18, 19, 20, 21, 22, 23, 24]. However, the presence of a conductive core modifies the system’s response relative to a purely resistive body, and this modification is most apparent when examining magnetopause boundary motion and precipitation patterns as a function of time. References: [1] Smith et al. (2012) Science, [2] Hauck et al. (2013) JGR: Planets, [3] Anderson et al. (2011) Science, [4] Johnson et al. (2012) JGR: Planets, [5] He et al. (2017) JGR: Space Physics, [6] Korth et al. (2017) GRL, [7] Nevsky et al. (2024) Universe, [8] Slavin et al. (2009) Science, [9] Glassmeier et al. (2007) SSR, [10] Johnson et al. (2016) GRL, [11] Katsura et al. (2021) Icarus, [12] Jia et al. (2015) JGR: Space Physics, [13] Grosser et al. (2004) PSS, [14] Heyner et al. (2016) JGR: Space Physics, [15] Shi et al. (2025) Icarus, [16] Fatemi et al. (2017) J. Phys.: Conf. Ser., [17] Fatemi et al. (2018) A&A, [18] Slavin et al. (2014) JGR: Space Physics, [19] Fatemi et al. (2020) JGR: Space Physics, [20] Varela et al. (2015) PSS, [21] Exner et al. (2024) JGR: Space Physics, [22] Jia et al. (2019) JGR: Space Physics, [23] Guo et al. (2023) JGR: Planets, [24] Glebe et al. (2026) JGR: Space Physics.

The precipitation of charged particles onto Mercury's surface is a key process in the Hermean system, as it is one of the main loss mechanisms for magnetospheric particles as well as results in surface sputtering, an important source of charged and neutral particles in the Hermean magnetosphere. These processes strongly depend on the properties of the precipitating particles, in particular their energy, which controls not only how they interact with the surface but also where precipitation occurs.In this study, we use single-particle tracing numerical simulations to investigate the effect of the global magnetospheric electric field on precipitating particle properties. Ion measurements made by the Miniature Ion Precipitation Analyzer (MIPA) instrument onboard BepiColombo during its second Mercury flyby are used as inputs for our model.Using these observed ion distributions, we show that the inclusion of an electric field in the simulations generates a dawn-dusk asymmetry in both the energies of the precipitated ions and the precipitating flux. Furthermore, we demonstrated that the presence of an electric field allows for increased dayside ion precipitation, created by low-energy ions transported towards the dayside by the E x B drift. This mechanism produces low-latitude and dayside precipitation regardless of upstream conditions, e.g. solar wind dynamic pressure or interplanetary magnetic field (IMF) orientation, which contrasts with results from hybrid simulations.The charged particles traced in this work will eventually interact with Mercury's surface, inducing e.g. the release of Energetic Neutral Atom (ENA) stripped from the surface which could be monitored by the ENA instrument onboard BepiColombo. This study therefore illustrates that the combined analysis of ion and ENA measurements together with particle tracing simulations provides a powerful framework to better understand the plasma circulation in Mercury's magnetosphere.

Mercury’s exosphere is a complex and dynamic environment, continuously refilled and depleted by processes acting both on the surface and within the planet environment (Milillo et al., 2005). The release processes are strongly driven by external factors such as the solar irradiance, the solar wind and the micrometeoroid flux precipitating onto the surface. In order to investigate the dynamics of planet environment, since its discovery in 1985 (Potter et al., 1985), Mercury’s exospheric Na has been the most studied species both from ground-based and from space observations. However, most of these studies based the analysis on single datasets, leaving a comprehensive and simultaneous characterization of space- and ground-based observations still unrealized. In this framework, we investigate and quantify the processes driving the Mercury's Na exosphere by merging complementary information from both the Telescope Heliographique pour l’Etude du Magnetisme et des Instabilites Solaires (THEMIS) ground-based telescope observations and the NASA/MESSENGER space-based observations.Since 2007, THEMIS has provided a large statistical database of bidimensional Na maps, revealing generally non-uniform emission patterns and showing a typical Na emission feature with two peaks located in the region of the magnetic cusp’s footprint (Fig. 1) (Potter et al., 2006; Leblanc et al., 2008; Mangano et al., 2015). These observations are complementary to in-situ measurements from MESSENGER mission (2008–2015), where data collected by the Ultraviolet and Visible Spectrometer (UVVS), part of the Mercury Atmospheric and Surface Composition Spectrometer (MASCS) onboard the spacecraft (McClintock and Lankton, 2007; Solomon et al., 2007), provided high-resolution altitude profiles and characterized the seasonal and annual behavior of the Na exosphere (Fig.2) (Cassidi et al., 2015). THEMIS and MESSENGER measurements are partially overlapped in time. By identifying high-quality joint observation windows from 2011 to 2013, while the telescope observes Mercury’s entire disk and surrounding space, providing the global view of the general status of the exosphere, the in-situ UVVS observations provide column density measurements at high resolution from which the altitude profile can be reconstructed. The derived scale heights allow estimating the release energy of exospheric particles and identifying the exospheric component distributions both in local time and latitude, as well as the dynamic evolution of the Na tail along the planetary orbit.To investigate these joint observations, we use the exosphere generation model developed at the Institute for Space Astrophysics and Planetology (IAPS) (Mura et al. 2007) that includes all the major sources of generation and loss processes responsible for the generation of Mercury's exosphere: Photon-Stimulated Desorption (PSD), which populates high-altitude layers; Thermal Desorption (TD), concentrated around the sub-solar point; and the combined effects of Ion Sputtering and Micrometeoroid Impact Vaporization (MMIV). Starting from the MESSENGER observation geometries during each orbital pass, we reconstruct the Na altitude profile along the line of sight of the instruments aboard the spacecraft, constraining the model parameter that best reproduce the observations. The same model parameters are applied to reproduce the THEMIS Na maps by integrating simulated densities along the lines of sight of the telescope (Fig. 3).Thanks to this work, it is possible to directly compare the different observations and derive additional information on the 3D structure of the Na exosphere. Such comprehensive and original analysis of the variability of Mercury’s Na exosphere represents an unprecedented step forward to better understand what drives the dynamics of planetary exosphere. Futhermore, this work will support the upcoming investigations by the ESA/JAXA BepiColombo mission, providing the framework necessary to interpret future multi-instrumental measurements of Mercury’s dynamic environment. Figure 1: Examples of Na emission patterns identified in the Hermean exosphere seen during a set of images collected from 2009 to 2013 with the THEMIS solar telescope (Mangano et al. 2015).Figure 2: Image of the UVVS observation geometry during the MESSENGER flybys of MercuryFigure 3: Comparison of observation geometries for Mercury’s Na exosphere (example for 7 June 2012): THEMIS ground-based configuration shows lines of sight from the Earth-based observer’s perspective (left); MESSENGER in-situ geometry illustrates the lines of sight during orbital passes (right)- Cassidy, T. A., Merkel, A. W., Burger, M. H., Sarantos, M., Killen, R. M., McClintock, W. E., & Vervack Jr, R. J., 2015. Mercury’s seasonal sodium exosphere: MESSENGER orbital observations. Icarus, 248, 547-559.- Leblanc, F., Doressoundiram, A., Schneider, N., Mangano, V., L.pez Ariste, A., Lemen, C., Gelly, B., Barbieri, C., Cremonese, G., 2008, High latitude peaks in Mercury’s sodium exosphere: spectral signature using THEMIS solar telescope. Geophys. Res.Lett.35,18- Mangano, V., Massetti, S., Milillo, A., Plainaki, C., Orsini, S., Rispoli, R., Leblanc, F., 2015, THEMIS Na exosphere observations of Mercury and their correlation with in-situ magnetic field measurements by MESSENGER, Planet. Space Sci. 115, 102–109- McClintock, W.E., Lankton, M.R. (2007). The Mercury atmospheric and surface composition spectrometer for the MESSENGER mission. Space Sci. Rev. 131 (1-4), 481-521- Milillo, A., Orsini, S., Wurz, P., Delcourt, D., Kallio, E., Killen, R.M., Lammer, H., Massetti, S., Mura, A., Barabash , S., Cremonese, G., Daglis, I.A.,De Angelis, E., Di Lellis, A.M., Livi, S., Mangano, V., Torkar, K., 2005, Surface–exosphere–magnetosphere system of Mercury. SpaceSci.Rev. 117(3), 397–444- Mura, A., Milillo, A., Orsini, S., Massetti, S., 2007. Numerical and...

The surface-bounded sodium exospheres of Mercury and the Moon are easily observed from the ground thanks to the fact that sodium is one of the brightest observable elements. Sodium act as crucial tracer for understanding exospheric dynamics in the inner Solar System.While both celestial bodies share common release mechanisms—such as photon-stimulated desorption, micrometeoroid impact vaporization, and ion sputtering—their exospheric behaviors differ significantly due to their unique orbital parameters and space environments.In fact, Mercury’s highly eccentric orbit, 3:2 spin-orbit resonance, and intrinsic global magnetic field lead to pronounced seasonal variations, longitudinal "cold-pole" accumulations, and intense interactions with the solar wind at the magnetic cusps.In contrast, the Moon is tidally locked and lacks a global magnetic field, meaning its exospheric variations are more heavily influenced by its periodic passage through Earth's protective magnetotail, localized crustal magnetic anomalies, and transient meteoroid streams.Furthermore, while both bodies exhibit massive, comet-like alkali tails driven by solar radiation pressure, Mercury's tail experiences extreme seasonal modulation based on its orbital true anomaly, whereas the lunar tail is uniquely observed via gravitational focusing by the Earth during the New Moon phase.Comparing these two distinct environments highlights the complex interplay among surface processes, magnetospheric interactions, and space weather effects, and provide important hints on how the comparative planetology may act to disentangle processes and ambient to improve their comprehension.We show how our long experience in the study of Mercury’s sodium exosphere morphology and dynamics can contribute a better understanding of the Moon’s exosphere and, more broadly, of the exospheres of airless bodies throughout the Solar System.

We present a global hybrid-particle simulation study of Mercury's magnetosphere and plasma environment during BepiColombo's first three Mercury flybys, or swing-bys, (MFB1–3). We use the high-performance simulation code RHybrid (paRallel Hybrid) to model the interaction between the solar wind and the Hermean magnetosphere[1-3]. In the hybrid approach, ions are treated as macroscopic particle clouds (macroparticles), resolving the kinetic motion of ions via the Lorentz force. Electrons are described implicitly as a charge-neutralising, inertialess fluid governed by Ohm's law and a polytropic (adiabatic) closure. The kinetic ion dynamics are coupled to the evolution of the magnetic field through Ampère's and Faraday's laws. This approach allows ion velocity distributions to evolve self-consistently, capturing wave-particle interactions, finite Larmor radius effects, and other ion-kinetic processes. Mercury's surface is modelled as a particle-absorbing inner boundary, while the crust–mantle region is represented as a resistive spherical shell atop an ideally conducting core. The planetary magnetic field is modelled as a dipole offset northward from the planet's centre. The solar wind ion species include protons and alpha particles, and Na+ ions are produced by photoionisation of a neutral sodium exosphere profile.For each flyby, we perform a dedicated simulation run using stationary upstream solar wind, interplanetary magnetic field (IMF), and photoionisation conditions representative of the observed environment [3-9]. Mercury was near perihelion during MFB3, while MFB1 and MFB2 occurred when the planet was about halfway between perihelion and aphelion. Under the adopted conditions, the solar wind was densest and the IMF strongest during MFB3, the solar wind speed was highest during MFB2, and the Alfvén Mach number was highest during MFB1. From these simulations, we generate virtual spacecraft observations along the MFB1-3 trajectories and compare boundary crossings, particle populations, electric and magnetic field properties, and wave activity in different regions of Mercury's plasma environment. The study aims to investigate how variations in upstream solar wind and IMF conditions, and exospheric ion loading influence Mercury's magnetospheric structure across the three encounters. We will discuss the comparative analysis of the model runs with BepiColombo and MESSENGER observations, as well as other modelling efforts.[1] https://github.com/fmihpc/rhybrid, https://planets.fmi.fi[2] Jarvinen et al (2020), Ultra-low frequency waves in the ion foreshock of Mercury: A global hybrid modeling study, Mon. Notices Royal Astron. Soc., 491, 3, 4147-4161, doi:10.1093/mnras/stz3257[3] Kallio et al. (2022), Ultra-low frequency waves in the Hermean magnetosphere: On the role of the morphology of the magnetic field and the foreshock, Geophys. Res. Lett. 49, 24, doi:10.1029/2022GL101850[4] Alberti et al. (2023), High-energy particle enhancements in the solar wind upstream Mercury during the first BepiColombo flyby: SERENA/PICAM and MPO-MAG observations, A&A 669, A35, 10.1051/0004-6361/202244662[5] Exner et al. (2020), Influence of Mercury's exosphere on the structure of the magnetosphere. Journal of Geophysical Research: Space Physics, 125, e2019JA027691, 10.1029/2019JA027691[6] Orsini et al. (2022), Inner southern magnetosphere observation of Mercury via SERENA ion sensors in BepiColombo mission, Nat. Commun. 13:7390, 10.1038/s41467-022-34988-x[7] Rojo et al. (2025), Characterization of the solar wind context during the third Mercury flyby of BepiColombo A&A, 698, A221, 10.1051/0004-6361/202553870[8] Teubenbacher et al. (2024), Solar wind entry into Mercury’s magnetosphere: Simulation results for the second swingby of BepiColombo, A&A, 681, A98, 10.1051/0004-6361/202347789[9] Teubenbacher et al. (2025), Hybrid modeling of Mercury’s magnetosphere: Assessing accuracy in ion counting statistics, A&A, 698, A12, 10.1051/0004-6361/202453452

Mercury's lack of atmosphere allows magnetospheric and solar wind plasma direct access to the surface through precipitation, altering the surface and plasma composition. In this way, its space plasma environment interactions resemble that of the Moon. Lunar measurements show that such exposed surfaces are subject to electrostatic charging, where photoelectrons generated on the dayside creating a positive potential of a few V, while electron transport to the nightside induces a negative surface potential up to -100s V. Given Mercury's similar surface properties and long diurnal cycle, we expect a similar pattern in surface potential; indeed, the nightside potential is likely even more negative at Mercury than at the Moon due to high energy electron precipitation.Lunar Prospector electron reflectrometry data indicates that this surface potential can be estimated remotely from pitch angle distributions of reflected particles, as the potential adds an energy-dependent term to the particle loss cone [1]. For positively charged ions and a negative surface potential, the size of the upward loss cone becomes inversely proportional to ion energy and can include all pitch angles < 90 degrees for sufficiently negative potentials. Thus, this effect has important implications for the ability of Mercury's magnetic field to trap particles on the nightside. We examine nightside pitch angle distributions from the Miniature Ion Precipitation Analyzer (MIPA) in the SERENA package from BepiColombo's second Mercury flyby for evidence of this loss cone energy dependence and find a noticeably visible effect for upward-going ions. Lunar surface potential analysis shows a strong dependence on electron temperature [2], so we use Mercury Electron Analyzer (MEA) electron temperature measurements from MFB2 [3] as a basis for fitting the loss cone and find good agreement with the MIPA data. This case study shows that estimation of surface potential from ion measurements is indeed possible, and promises interesting results to come from the BepiColombo orbital phase.[1] Halekas, J. S., G. T. Delory, R. P. Lin, T. J. Stubbs, and W. M. Farrell. “Lunar Prospector Observations of the Electrostatic Potential of the Lunar Surface and Its Response to Incident Currents.” JGR: Space Physics 113, no. A9 (2008). https://doi.org/10.1029/2008JA013194.[2] Stubbs, T. J., W. M. Farrell, J. S. Halekas, et al. “Dependence of Lunar Surface Charging on Solar Wind Plasma Conditions and Solar Irradiation.” P&SS 90 (January 2014): 10–27. https://doi.org/10.1016/j.pss.2013.07.008.[3] Rojo, M., N. André, S. Aizawa, et al. “Structure and Dynamics of the Hermean Magnetosphere Revealed by Electron Observations from the Mercury Electron Analyzer after the First Three Mercury Flybys of BepiColombo.” A&A 687 (July 2024): A243. https://doi.org/10.1051/0004-6361/202449450.

Mercury’s exosphere has been extensively investigated through ground-based observations. Since 2007, the THEMIS solar telescope was used to conduct a long-term monitoring campaign of the sodium (Na) exosphere, enabling detailed studies of its morphology and variability at different timescales driven by interactions between Mercury’s surface, intrinsic magnetic field, neutral and ionized environment, and the solar wind and radiation. They allowed to investigate the contribution of the different source processes as well as the statistical frequency of the different morphological patterns, and the rapid change from one to another at the sudden change of interplanetary magnetic field conditions.In 2022, 2023, and 2024, we carried out targeted observations of Mercury’s sodium exosphere in the days surrounding three of the six flybys performed by the ESA–JAXA BepiColombo spacecraft around the planet. The objective was to provide a global view of the exospheric morphology and dynamics during the spacecraft’s close-approach phase, when also several in-situ instruments were simultaneously operating. These measurements offered crucial information on the planetary environment, including magnetic fields, as well as ion, electron, and neutral populations across a wide range of energies, and radiation.We present and compare the three exospheric configurations observed in the days around these three flybys, highlighting their morphological and dynamical similarities and differences.