TP8
Key Themes:
• Atomistic Modeling: Utilizing atomistic approaches to provide insights and possible validations for larger-scale processes.
• Parameter Identification: Determining key parameters for surface-bounded exospheres and surface interactions at varying scales will inform further atomistic research and guide observational and laboratory experiments.
• Surface–Exosphere Linkages: Explaining the connections between surface-bounded exospheres and their respective surfaces, including compositional and topographic effects.
• Comparative Investigations: Encouraging comparative studies across different planetary bodies (including icy moons) to enhance our knowledge of underlying processes.
By addressing these key themes, the session explicitly links micro-scale processes to macro-scale dynamics, thereby bridging scales and enriching our integrated understanding of exosphere–surface interactions. We particularly encourage early career scientists to submit an abstract for an oral presentation.
Session assets
Introduction
Mercury’s surface undergoes extreme diurnal and annual temperature variations driven by high-intensity insolation, radiative loss due to its lack of atmosphere, its spin-orbit resonance, and eccentric orbit. These thermal variations have proven important for the regolith production [1], caused by thermal cracking on atmosphereless planetary bodies. In this study, we aim to examine the solar radiation effects on Mercury’s surface regolith.
Measurements
We simulate the power density of the solar radiation impinging on a planetary surface at Mercury’s perihelion distance of 0.31 au using the Space and High-Irradiance Near-Sun Simulator (SHINeS) [2] at Luleå University of Technology to study the spectral response of terrestrial analogue materials. The analogues used include basalt, gabbro, andesite, anorthosite, diorite, and boninite, and they were selected for their close geochemical properties to Mercury’s surface. Notably, boninite is currently considered one of the closest terrestrial hermean analogues. The samples were ground and sieved to <75 μm to simulate the fine-grain regolith of Mercury [3]. The reflectance spectra in the visible to near-infrared (VIS/NIR) range (0.44 - 2.5 μm) were acquired pre- and post-irradiation to evaluate spectral changes.
Results
The experimental results (Figure 1) show a decrease in reflectance in all samples across the VIS and NIR spectra, except gabbro, which increases brightness in the NIR above 1470 nm. Darkening and reddening were more pronounced in samples with higher Fe content, such as boninite and basalt, while gabbro, despite its moderate Fe content, remained spectrally stable. We attribute this discrepancy to the dependence of solar radiation absorption on the samples’ albedo. The less reflective samples absorb more energy.
Figure 1: The reflectance spectra in the visible and near-infrared regions for the six samples - basalt (BAS), andesite (AND), anorthosite (ANO), boninite (BON), diorite (DIO), and gabbro (GAB) – before (blue line) and after (red line) irradiation at 1.40 W/cm2. Most samples exhibit a reduction in brightness in the VIS and NIR spectra. Gabbro presents an increase of brightness in the NIR above 1470 nm. In the NIR, boninite shows a significant drop in brightness along with changes in its absorption band depths [Latsia et al. 2025, in review].
References
[1] Molaro J. and Byrne, S. 2012. Rates of temperature change of airless landscapes and implications for thermal stress weathering. Journal of Geophysical Research: Planets, 117(E10).
[2] Tsirvoulis, G., Granvik, M. and Toliou, A., 2022. SHINeS: Space and High-Irradiance Near-Sun Simulator. Planetary and Space Science, 217, p.105490.
[3] Domingue, D.L., Chapman, C.R., Killen, R.M., Zurbuchen, T.H., Gilbert, J.A., Sarantos, M., Benna, M., Slavin, J.A., Schriver, D., Trávníček, P.M. and Orlando, T.M., 2014. Mercury’s weather-beaten surface: Understanding Mercury in the context of lunar and asteroidal space weathering studies. Space Science Reviews, 181, pp.121-214.
[4] Hunt, G.R. and Ashley, R.P., 1979. Spectra of altered rocks in the visible and near infrared. Economic Geology, 74(7), pp.1613-1629.
[5] Maturilli, A., Helbert, J., John, J.M.S., Head III, J.W., Vaughan, W.M., D'Amore, M., Gottschalk, M. and Ferrari, S., 2014. Komatiites as Mercury surface analogues: Spectral measurements at PEL. Earth and Planetary Science Letters, 398, pp.58-65
How to cite: Latsia, N., Tsirvoulis, G., Kaufmann, E., Haack, D., Granvik, M., and Hagermann, A.: Experimental investigation of solar radiation effects on Mercury’s surface regolith, EPSC-DPS Joint Meeting 2025, Helsinki, Finland, 7–13 Sep 2025, EPSC-DPS2025-614, https://doi.org/10.5194/epsc-dps2025-614, 2025.
Introduction
It is now generally accepted that water and other volatiles exist on the Moon [1, 2], though their distribution and abundance are still poorly-constrained. Much work has been put into understanding how lunar volatiles are transported from their delivery sites to the polar cold traps where they have been detected. These models can generally be divided into two regimes: collisional transport through a transient post-impact atmosphere [3] and ballistic transport through a non-collisional surface-bounded exosphere [e.g. 4, 5, 6]. Transient atmospheres allow for rapid but episodic delivery of volatiles to cold traps, while ballistic transport permits a much slower but potentially continuous delivery.
Given the exponential dependence of the residence time on temperature, there has been significant focus on characterizing the Moon’s thermal environment, particularly at scales below the resolution of existing orbital datasets [e.g. 3, 7]. However, comparatively little attention has been dedicated to another key parameter: the desorption activation energy. The activation energy (Ea) is as significant to a molecule’s time adsorbed to the surface as the temperature is, with larger values leading to longer residence times:
Most ballistic transport models use a single value for the activation energy. This is a problematic assumption for several reasons. Temperature-programmed desorption (TPD) measurements of water desorption from Apollo samples have indicated that a single surface can have a broad range of activation energies [8, 9]. Additionally, the lunar surface does not have a uniform composition, with the most significant compositional dichotomy occurring between the maria and the highlands [10]. There is no reason to assume that these different surfaces would have the same (or even similar) activation energies. This allows for the potential of a positional dependence on the ability of water and other species to desorb from the surface, which may challenge the idea that ballistic transport results in a more-or-less uniform delivery to all cold traps.
Furthermore, the specific value (or range of values) of the activation energy is not well-understood, particularly for non-water volatiles. The applicability of a frequently-cited value of 0.415 eV for water is questionable, as this value was derived for water molecules sublimating from a water-ice substrate [11]. Except for limited areas within the polar cold traps, this is unlikely to be representative of real-world conditions. Even a small (~10%) increase in the activation energy has the potential to measurably affect transport behaviour, particularly when small-scale surface roughness is considered due to the greater dependence of residence time on temperature for larger activation energies [3].
Our goal is to highlight the importance of developing a more comprehensive understanding of activation energies for lunar volatiles, whether through experimental work (e.g. TPD) or models (e.g. molecular dynamics).
Methods
We use a standard ballistic transport model adapted from the one presented in Kloos et al. [6]. Molecules desorb from the surface in a random direction with a speed chosen from the Maxwell-Boltzmann distribution for the residence site’s surface temperature. For simplicity, we do not include the effects of small-scale roughness. Molecules hop across the surface until they land in a cold trap or are photolyzed. The activation energy is modeled both as a single value and as a distribution of values, following the form laid out by Schöghofer [12].
Results
An overview of the relationship between surface temperature, desorption activation energy, and surface residence time is presented in Figure 1. As the activation energy increases, so does the temperature at which the residence time is a significant fraction of a lunar day. This effect is more pronounced at lower temperatures, suggesting that transport near the lunar poles may be particularly affected by the choice of activation energy, particularly given the seasonally-shadowed regions that create complex and time-variable areas of sustained low temperatures [6].
We plan to present results examining a wide range of parameter space, including uniform and non-uniform surface compositions and the relative importance of single-value activation energies versus a broad distribution of energies.
Figure 1. The interrelated effects of desorption activation energy and surface temperature on the surface residence time for water molecules. The horizontal dotted line in each panel represents a residence time of half a lunar day. The vertical dotted line in the left panel indicates the typical maximum temperature of the Moon’s permanently-shadowed regions.
References
[1] Colaprete, A., et al. (2010). Detection of water in the LCROSS impact plume. Science, 330(6003):463–468.
[2] Li, S., et al. (2018). Direct evidence of surface exposed water ice in the lunar polar regions. Proceedings of the National Academy of Science, 115(36): 8907–8912.
[3] Prem, P., et al. (2018). The influence of surface roughness on volatile transport on the Moon. Icarus, 299:31–45.
[4] Schörghofer, N. (2014). Migration calculations for water in the exosphere of the Moon: Dusk-dawn asymmetry, heterogeneous trapping, and D/H fractionation. Geophysical Research Letters, 41(14):4888–4893.
[5] Moores, J. E. (2016). Lunar water migration in the interval between large impacts: Heterogeneous delivery to Permanently Shadowed Regions, fractionation, and diffusive barriers. Journal of Geophysical Research: Planets, 121(1):46–60.
[6] Kloos, J. L., et al. (2021). Illumination conditions within permanently shadowed regions at the lunar poles: Implications for in-situ passive remote sensing. Acta Astronautica, 178:432–451.
[7] Hayes, C. W., et al. (2024). Topography-enhanced ultra-cold trapping at the LCROSS impact site. Journal of Geophysical Research: Planets, 129(7):e2023JE007925.
[8] Poston, M. J., et al. (2015). Temperature programmed desorption studies of water interactions with Apollo lunar samples 12001 and 72501. Icarus, 255:24–29.
[9] Jones, B. M., et al. (2020). Investigation of water interactions with Apollo lunar regolith grains. Journal of Geophysical Research: Planets, 125(6):e06147.
[10] Yang, C., et al. (2023). Comprehensive mapping of lunar surface chemistry by adding Chang'e-5 samples with deep learning. Nature Communications, 14:7554.
[11] Sandford, S. A. & Allamandola, L. J. (1988). The condensation and vaporization behavior of H2O: CO ices and implications for interstellar grains and cometary activity. Icarus, 76(2):201–224.
[12] Schörghofer, N. (2023). Adsorption kinetics of water and argon on lunar grains. The Planetary Science Journal, 4(9):164.
How to cite: Hayes, C. and Moores, J.: Characterizing the Importance of Desorption Activation Energies on Delivery Rates of Volatiles to the Lunar Cold Traps, EPSC-DPS Joint Meeting 2025, Helsinki, Finland, 7–13 Sep 2025, EPSC-DPS2025-83, https://doi.org/10.5194/epsc-dps2025-83, 2025.
Introduction.
Traditionally, models of surface-bounded exospheres assumed a perfectly spherical planetary surface. Neglecting the surface roughness and topography simplifies the surface-exosphere interaction and reduces the computational complexity and effort. One of the consequences of this simplification can be seen in the surface temperatures, which are often included as a simple analytical formula based solely on the subsolar location, or as a look-up function based on remote sensing data (e.g., Butler, 1997; Crider an Vondrak, 2002; Killen et al., 2019). These approaches do not account for the surface variability and local shadowing, which causes colder surfaces and hotter slopes tilted towards the Sun.
Recent studies have begun to address surface roughness in exosphere modelling: Prem et al. (2018) introduced small‐scale shadows through Gaussian roughness in water‐exosphere simulations, finding enhanced water retention; Grava (2023) used altitudes from LOLA and temperature snapshots from Diviner to find that topography and its resulting change in surface temperatures plays an important role for the exosphere; Laferriere et al. (2025) investigated sloped temperature distributions for OH/H₂O, reporting locally colder surfaces and higher adsorbed concentrations. Though, their low particle counts highlighted a need for more numerically efficient approaches. In this work, we present a new and efficient approach to including relevant surface slopes in a lunar exosphere Monte Carlo simulation to model noble-gas particle density distributions and investigate the effects of slopes. This study aims to quantify how realistic slope distributions based on the LOLA dataset influence noble gas distributions in the lunar exosphere.
Methods.
High‐resolution LOLA topography (512 px/°) was processed into cumulative density functions (CDFs) of slopes for 1°x360° latitude strips, depicted in Figure 1. Each line in this plot represents the CDF of a single strip, showing that topographic slope distributions on meter scales across the entire Moon’s surface are relatively similar, with only some exceptions close to the poles. In our model, a new slope is drawn from the CDF at the corresponding latitudinal position at every surface interaction, including topography purely probabilistically and, thus, more efficiently.
Figure 1. Cumulative distribution functions of surface slopes from 1°x360° latitude strips on the Moon, based on the LOLA dataset. Each line represents one strip, with more than 90 % of slopes falling below 20°. The maximum observed slope approaches 80°.
Surface temperatures were computed using the model of Butler (1997), as adapted by Crider & Vondrak (2002) and Killen et al. (2019), with the local solar incidence angle adjusted for slope magnitude and azimuth, based on Duffie and Beckman (2013). This resulted in a distribution of surface temperatures at every subsolar location linked directly to the slope distribution. Since the mean temperatures of these distributions are lower than their reference, the no-slope temperature of the analytical model at the same location, an additional temperature offset (-20K to +60K) was included in the simulation to keep the global thermal environment as similar as possible. Figure 2 shows this temperature offset at each solar incidence angle (black markers) and its fitting (orange line), done with a 20-step multilinear, exponentially spaced model.
Figure 2. Difference between sloped‐surface and flat‐surface temperatures as a function of solar incidence angle, and the 20‐step multilinear offset fit.
This topography and surface temperature model was included in our particle-tracking code, extending on previous work described by Smolka et al. (2023) for three-dimensional exospheres. Release velocities were also rotated according to the slope and its azimuth, introducing two main changes: the geometrical, and the thermal influence of slopes on exosphere densities and dynamics.
Preliminary and Expected Results.
The first results show that the probabilistic approach of surface topography can be included in a Monte Carlo simulation without any major drawbacks regarding numerical efficiency. Surface number densities of Helium, see Figure 3, show increases up to ~25% along the terminator regions. This is strongly connected to the temperature model, which has the biggest spread close to solar incidence angles of 90°, see Figure 2. Preliminary studies of Neon showed similar effects, though, with lower density increases, consistent with cooler temperatures and more surface‐hugging trajectories. The effects on the Argon densities are expected to be stronger, compared to both Helium and Neon, due to adsorption on cold surfaces using the presented topography model.
Figure 3. Heatmap of the percentage increase in steady‐state Helium surface density when including slopes, relative to a flat‐surface simulation.
Conclusion and Future Work.
The planned validation against LADEE and LAMP exospheric density measurements will assess implications for satellite data interpretation. We will extend the refined temperature model to hydrogen‐based species to quantify topography effects on the lunar water cycle. In the long term, the incorporation of advanced surface temperature models (e.g., Hurley et al., 2015) and Diviner data will bridge the gap to more realistic simulations.
This research demonstrates not only that including realistic data-based topography in exosphere simulations is computationally feasible but also highlights its necessity for capturing lunar exospheric dynamics.
References.
Butler (1997). Journal of Geophysical Research, 102(E8), 19283–19291.
Crider and Vondrak (2002). Advances in Space Research, 30(8), 1869–1874.
Duffie and Beckman (2013). Solar Engineering of Thermal Processes (4th ed.). John Wiley.
Hurley et al. (2015). Icarus, 255, 159–163.
Killen et al. (2019). Icarus, 329, 246–250.
Prem et al. (2018). Icarus, 299, 31–45.
Grava (2023). Dust, Atmosphere, and Plasma Environment of the Moon and Small Bodies (DAP-2023).
Smolka et al. (2023). Icarus, 397, 115508.
Laferriere et al. (2025). Journal of Geophysical Research: Planets, 130(4).
How to cite: Peschel, A., Grava, C., Breternitz, J., and Reiss, P.: Probabilistic Modelling of Lunar Topography for Improved Exosphere Simulations, EPSC-DPS Joint Meeting 2025, Helsinki, Finland, 7–13 Sep 2025, EPSC-DPS2025-451, https://doi.org/10.5194/epsc-dps2025-451, 2025.
Introduction
Sputtering in planetary science occurs as the solar-wind (SW)—a stream of energetic ions emitted from the Sun—impacts an airless body, ejecting atoms from its surface [1,2]. This process alongside micrometeorite impacts, photo-stimulated desorption, and thermal desorption contribute to the formation of planets’ exospheres [3–5]. While various spacecraft can detect exospheric species such as MESSENGER, BepiColombo, LADEE, and CHACE-2, they cannot discern the respective contributions of the mentioned processes and thus a strong theoretical understanding of sputtering is needed to quantify its influence on the exosphere [6–9].
The sputtering yield is well-studied. In contrast, the angular distribution of ejecta has been given significantly less attention, its treatment being particularly sparse in cases relevant to planetary science. As such, sputtering models that consider the angular distribution of ejecta often assume isotropy. Here, we present a theoretical study quantifying anisotropy in the angular distribution of ejecta for SW-induced sputtering cases, helping advance the understanding of sputtering’s contribution to exosphere formation. Further, we compare the results to a common experimental case since experiments often employ heavier, higher energy ions to leverage the enhanced mass detection consequent of a greater sputter yield. These experimental results must then be scaled to inform SW-induced sputtering and, as such, unique behaviors occurring for lower mass impactor cases may be overlooked. Finally, following quantification, we consider the relative contributions from four ejecta-types demonstrated in Fig. 1, an approach enabling us to understand the underlying behavior leading to anisotropy differences between the different cases considered.
Fig. 1: An incident ion (red) impacts a target, collides with atoms within, and exits as a reflected ion, triggering four ejecta-types in the process (blue), from left to right: ion-in SKAs, ion-in PKAs, ion-out PKAs, and ion-out SKAs.
Methodology
To simulate sputtering, we utilized the software SDTrimSP which follows the binary collision approximation (BCA) model where sputtering occurs through a sequence of independent collisions within a material prior to the ejection of an atom [1,2]. While both electronic and collisional effects occur in the sputtering process, the latter dominate at energies below 100keV amu-1 and thus we consider collisional sputtering exclusively [2].
We selected 1 keV ionized Hydrogen (H+) and 4 keV ionized Helium (He++) to emulate the SW, while 20 keV ionized Krypton (Kr+) was employed given its prevalence in experimental studies. For the target surface, silica (SiO2) was selected given its prominence in both the lunar and Mercurian surfaces and recurrent usage in experiments [10–12]. We simulated ion incidence angles between 0° and 85° (measured from the surface normal) while ejecta were interpreted as a function of polar and azimuthal angles, ranging from 0° to 90° and 0° to 180°, respectively. The scenario is illustrated in Fig. 2.
Fig. 2: An incident ion impacts a target substrate at an incidence angle, θi, sputtering an atom as a function of polar (θs) and azimuthal (φs) angles within the depicted quarter-sphere.
Results
Forward-backward anisotropy exists when a greater percentage of atoms are sputtered at azimuthal angles between 0° and 90° than 90° and 180°. While the azimuthal distribution of ejecta is isotropic at normal incidence, anisotropy emerges as the ion incidence angle is varied. Noticeable differences in anisotropies between ion cases arise as the ion’s incidence angle is made increasingly oblique, forward-backward anisotropy becoming most pronounced in the H+ case while developing more modestly in the He++ and Kr+ cases. Alternatively, to assess anisotropy in the polar distribution of ejecta we consider anisotropy occurring as a greater percentage of atoms are sputtered between 0° and 45° (“low” angles) than 45° and 90° (“high” angles). At normal incidence, low-angle anisotropy is prominent in all cases. With increasing incidence angle, the polar distribution of ejecta becomes more isotropic in the H+ case, slightly more anisotropic in the He++ case, while remaining relatively steady in the Kr+ case.
The divergence in the anisotropies witnessed in the H+ case from those occurring in the other two impactor cases considered can be explained by an interplay between the percentage contribution of specific ejecta-types and the extent to which they are forward and low-angle pronounced. On the one hand, the ejecta-types most readily sputtered forward and at high-angles are generally most prominent in the H+ case and on the other, individual ejecta-types in the H+ case typically have higher forward and lower low-angle sputtering percentages than those in the He++ and Kr+ cases.
Concluding Statement
The findings demonstrate that sputtering anisotropy varies significantly depending on the ion-target case considered. While anisotropies in the He++ and Kr+ cases are similar, there are clear differences in the case of H+ bombarding SiO2. Experimental cases using increased energies and masses are, therefore, likely underestimating the degree to which forward-backward anisotropy is present in SW-induced sputtering cases, while overestimating the extent of anisotropy in the polar distribution of ejecta. Accounting for these effects is essential when scaling experimental results to inform planetary sputtering.
How to cite: Clouter-Gergen, B., Morrissey, L., Bu, C., Mutzke, A., Verkercke, S., and Savin, D.: Solar Wind-Induced Sputtering: Investigating Anisotropy in the Angular Distribution of Ejecta using SDTrimSP, EPSC-DPS Joint Meeting 2025, Helsinki, Finland, 7–13 Sep 2025, EPSC-DPS2025-1193, https://doi.org/10.5194/epsc-dps2025-1193, 2025.