Simulating Vapour Plumes through Icy Moon Crevasses with DSMC

We present preliminary results of Direct Simulation Monte Carlo (DSMC) [1] simulations of plumes through icy moon crevasses. The subsurface liquid oceans of icy moons are some of the most fascinating places in the Solar System to search for habitable worlds (beyond Earth), and plumes venting from such crevasses present an opportunity to access fresh material from the subsurface reservoir, be that a water inclusion or a global subsurface ocean. The most promising targets are Jupiter’s icy moon Europa and Saturn’s icy moon Enceladus. For Europa, many studies have argued for the presence of plumes, though some recent developments revised the most prominent localized water plumes evidence [2]. For Enceladus, plumes were directly observed and characterized by many complementary instruments of the Cassini mission. In-situ sampling of the plume material by Cassini’s Ion and Neutral Mass spectrometer (INMS) and Cosmic Dust Analyzer (CDA) has been used to infer the presence of different species [3,4].

 

In the coming decade, ESA’s JUpiter Icy Moons Explorer (JUICE) [5] and NASA’s Europa Clipper [6] missions will conduct flybys of Europa and Jupiter’s other icy moons. Each spacecraft carries a mass spectrometer instrument (NIM on JUICE and MASPEX on Europa Clipper) which will analyze the chemical composition from gas samples of Europa’s exosphere. Plumes encountered directly by spacecraft can use the sampled composition to infer subsurface reservoir properties. If plumes are not directly encountered, exospheric simulations both with and without included plume modeules may help to support or dispel their presence via indirect evidence attributable to plume sources. ESA’s L4 mission plan for Enceladus further emphasizes the need to properly understand how plumes and plume compositions reflect on different subsurface water reservoir conditions.

 

To this end, we model the flow of gaseous plumes through the crevasse of icy moons with the DSMC method. We utilize the state-of-the-art open-source Stochastic PArallel Rarefied-gas Time-accurate Analyzer (SPARTA) code [7,8]. Our aim is a simulation with multiple species, with chemical composition inputs from chemical equilibrium models of the subsurface ocean [9] and with outputs being fed into an exosphere model. But first, we must focus on the fundamental parameters shaping the reactions of the different species and the crevasse walls. For example, in literature, a variety of Variable Soft Sphere (VSS) parametrizations for the collision cross section are encountered, even for the same species. But there exist general approaches to estimate VSS parameters, yielding collision property databases for a large number of species pairs (240 × 240) [10].

 

We compare SPARTA and the ultra-fast Statistical PARTicle Simulation (ultraSPARTS) [11] by running these two DSMC codes under consistent geometrical and physical initial conditions and comparing their outputs. Furthermore, we compare the impact of different collisional cross sections in both codes. In addition, we develop a Computational Fluid Dynamics (CFD) MacCormack solver for a version of the Navier-Stokes equations specifically for nozzled crevasses with a subsonic/supersonic flow transition, based on the approach by [12]. Using multiple types of simulations allows us to compare their results under simple, identical setups, as well as investigate the changes induced by complexities that can only be captured by one of the simulation types. For validation, we may compare our simulations to laboratory setups modelling plumes physically, e.g. the Crevasse Laboratory Analogues for Moons in the Plumes and Ices Simulation Chamber for Enceladus and other moonS (PISCES/CLAM) [13]. Our ultimate goal is to subject the modelled plumes and coupled codes to a reversal algorithm, to be able to constrain subsurface reservoirs from above-surface measurements.

 

Acknowledgement:

The authors acknowledge the financial support of the SNSF under SNSF starting grant 218336.

 

References:

[1] Graeme Bird. (1994). Molecular gas dynamics and the direct simulation of gas flows.

[2] Lorenz Roth, et al. (2026). Astronomy & Astrophysics, 709, A59. https://doi.org/10.1051/0004-6361/202659406 

[3] Brian Magee and J. Hunter Waite. (2017). Lunar and Planetary Science XLVIII, 2974. https://scixplorer.org/abs/2017LPI....48.2974M 

[4] Frank Postberg, et al. (2018). Nature 558, 564–568. https://doi.org/10.1038/s41586-018-0246-4

[5] Olivier Grasset, et al. (2013). Planetary and Space Science 78, 1-21. https://doi.org/10.1016/j.pss.2012.12.002

[6] Cynthia Phillips and Robert Pappalardo. (2014). Eos, Transactions AGU, 95(20), 165-167. https://doi.org/10.1002/2014EO200002

[7] Steve Plimpton, et al. (2019). Physics of Fluids 31, 086101. https://doi.org/10.1063/1.5108534

[8] http://sparta.github.io/

[9] Alizée Amsler Moulanier, et al. (2025). The Planetary Science Journal 6, 1. https://doi.org/10.3847/PSJ/ad9925

[10] Mitchell Gosma, et al. (2025) Physics of Fluids 37, 036156. https://doi.org/10.1063/5.0252567

[11] http://www.plasmati.com.tw/

[12] Nick van der Hijden, et al. (2024). Icarus 417, 116114. https://doi.org/10.1016/j.icarus.2024.116114

[13] Yaël Bourgeois and Stéphanie Cazaux. (2025). Planetary and Space Science 269, 106206. https://doi.org/10.1016/j.pss.2025.106206