GAS-SOLID INTERACTIONS IN THE ATMOSPHERE OF MARS AND
THEIR EFFECT ON METHANE AND TRACE-GAS EVOLUTION
J.E. Moores1
1
Centre for Research in Earth and Space Science, York University, Toronto, Canada (jmoores@yorku.ca)
Introduction: Recent observations of the evolution of trace gasses near the surface of Mars have
suggested a puzzle in atmospheric chemistry. Oxygen, carbon monoxide and methane all increase and
decrease over the year in a similar way (e.g. Trainer
et al., 2019), which is unexpected of such different
chemical species. Meanwhile, concentrations of methane in the near-surface atmosphere (Webster et al.,
2015; 2021) suggest more production of the gas than
can reasonably be destroyed by known mechanisms.
Several groups have posited a fast-destruction
mechanism to account for this imbalance in production and destruction of methane and the strange behavior of other gasses. Tantalizingly, many of the
proposed mechanisms rely on Gas-solid interactions
either with airborne dust or other surface materials.
For instance, Jensen et al. (2014) have proposed
chemisorption of methane onto grain surfaces.
Meanwhile Thøgersen et al (2019) have proposed
triboelectric destruction of methane between grains
and Atreya et al (2011;2006) and Delory et al (2006)
propose the chemical removal of methane by surface
oxidants. Furthermore, Atreya et al. (2021) and
Trainer et al. (2019) both suspect a surfaceatmosphere process to be responsible for the remarkable shifts seen in oxygen concentration and Korablev et al. (2021) have linked the uplift of salt-rich
surface materials to the surprising atmospheric concentration of HCl observed by TGO.
Regardless of the specific mechanism, a critical
step to determining the rate of change of the concentration of any gas will be the rate at which gas molecules encounter a surface where reactions can take
place. This rate of interaction will change throughout
the day and season, creating patterns in measured gas
concentrations. Those changes, in turn, will provide
evidence of gas-solid chemistry.
Grain Encounter Model: To estimate the rate of
grain encounters for any putative gas molecule, a
simple model is constructed.
First, all dust grains are assumed to be equidistant from one another, an assumption which is likely
true on average. Based on the dust loading described
by Moores et al. (2007) which found that 3 x 106
particles m-2 were required to explain typical optical
depths of 0.5 and assuming constant dust/gas mixing
ratio with height, we find that grains should be ~7
mm apart on average. If those grains are organized
as the vertices of a cube, it becomes possible to calculate how long a gas molecule would be expected to
take to diffuse from its start location to any one of
the vertices.
The path of the diffusing molecule is modelled as

an expanding sphere. As this sphere passes each dust
particle (located at each vertex), the probability of an
encounter between the gas molecule and the dust
particle is calculated based upon the fraction of the
surface of that sphere which is occupied by the
cross-section of the dust particle. For the purposes of
our model, we use the modal grain size which is
close to 0.5 microns in radius (Moores et al., 2007).
Once the expanding diffusion sphere touches the
most distant dust particle, the probabilities of an encounter with all dust particles are summed to give an
overall likelihood of an encounter with any of the
dust particles over the time required to diffuse to the
final grain. As the grid is a regular one, it becomes
possible to determine how long it would take to raise
the probability of an encounter to 1 by continuing to
expand the diffusion sphere. This is simply the time
to diffuse to the most distant grain divided by the
probability of an encounter along the way. It is necessary to obtain a probability of 1 because the diffusion front itself is an expression of the root mean
square distribution of the molecular distances. This
process is repeated at 1,000,000 different positions
of gas molecules interior to the cube with dust grains
as vertices. The average over the 1,000,000 gas molecules gives a good estimate of the time required for
any one gas molecule to have an encounter with a
dust grain.
The diffusion distance and the time required are
linked through the diffusivity. If the daytime is selected for calculation, the diffusion front is defined
by the eddy diffusivity, approximately 2000 m2 s-1,
as described in Taylor et al. (2007). This yields a
typical time between gas-dust interactions of 0.47
seconds. If nighttime is selected, we use the molecular diffusivity of 9.4 x 10-4 m2 s-1, consistent with
numerical model results for nighttime PBL behavior.
This yields a typical time between gas-dust interactions of 9.9 x 105 seconds or ~11 sols.
Modeling Fast Destruction Mechanisms with
the Grain Encounter Model: As an example, we
use the model to consider the evolution of methane
by deriving an atmospheric lifetime. When these
lifetimes are input to the model of Moores et al.
(2019), the effect is equivalent to adding a fast destruction mechanism. Even those molecules which
encounter a grain are not captured or destroyed with
perfect efficiency, so different lifetimes for methane
in the atmosphere and different amounts of background methane, perhaps derived from external
plumes are used in model runs. In all runs, the model
has been extended so that it continues past the time
of the SAM ingest and completes an entire diurnal

cycle. Figure 1 shows these runs using the sol 1709
SAM-TLS ingest from Moores et al. (2019).
The upper panel of figure 1 shows two model
runs with no background methane present. The black
curve shows the baseline case with a methane lifetime of 329 years (e.g. Atreya et al., 2006) while the
red curve shows an exceptionally fast destruction
mechanism with an exponential lifetime of only
3600 seconds (1 hour). Overnight, both curves are
identical as there are few gas-solid interactions and
therefore, the fast destruction mechanism does not
operate. As a result, the changes that occur in methane concentration are due entirely to the accumulation of methane into the near surface layer.

the scenario (red curve) with an exceptionally fast
destruction mechanism declines at dawn more quickly than does the baseline case (black curve), the separation between the two is only 0.05 ppbv (50 pptv)
at most. Not only is achieving this level of sensitivity
in a flight instrument incredibly difficult, but measurements would need to be acquired every few
minutes to quantify the effect. Instead, the rapid decline in methane levels observed is due almost entirely to the effect of dilution with methane-free air
above. As such, while micro-seepage can be quantified if background levels are low, the fast destruction
mechanism is nearly impossible to measure.
This situation changes if there is a plume event

Figure 1. Evolution of methane concentration
under different background levels and fast destruction mechanisms. In the top panel, there is no background methane and even fast destruction mechanisms that operate more quickly than any proposed
produce very little change in concentration. In the
bottom panel, the evolution of the concentration at 1
m overnight is the result of micro-seepage, whereas
the daytime evolution is due to the fast destruction
mechanism.

nearby which raises the total amount of background
methane in the atmosphere. The lower panel of Figure 1 shows this case with the baseline (black) compared to a 1-sol lifetime (blue), as suggested by
Lefèvre and Forget (2009), plus a 0.25 sol lifetime
(cyan) and a 3600 second lifetime (red). As with the
no background methane case, all curves are the same
overnight when the fast destruction mechanism does
not operate, but during the day each curve exhibits
different rates of change with the 3600 second case
achieving zero methane by the end of the day and
even a 1-sol lifetime yielding a detectable 0.2 ppbv
(200 pptv) decline. Note that in this model, the evo-

During the daytime, these curves are separated
slightly but remain nearly indistinguishable. While

lution of methane depends highly on the PBL height
as a rising PBL mixes down 1 ppbv of methane into
the near-surface. As such, the relative coarseness of
the bottom panel as compared to the top panel is the
result of the relative coarseness of the PBL model
inputs from Newman et al. (2017) as compared to
the 10 s timestep used in the model shown here.
As such, a surface spacecraft could measure the
micro-seepage rate on any day during the night,
whereas quantifying the fast destruction mechanism
would require daytime measurements and would
need a plume to be present to make useful observations. While such measurements of a declining
plume could in principle also be made from orbit,
having a surface station as well would provide a
unique perspective on the rate of change and would
be important to help determine not only the rate of
the fast destruction mechanism but whether it primarily operates near the surface or throughout the
column.
Gas-solid interactions can amplify many fast
destruction mechanisms. A fast destruction mechanism would allow for more methane destruction to
occur for the same measured concentration than
would otherwise be possible, allowing for more areas on Mars to emit methane via micro-seepage
(Moores et al., 2019) with the increase proportional
to the efficacy of the destruction mechanism. Such a
process must be periodically overwhelmed by larger
releases which are seen at all levels of the atmosphere, but also decay away quickly, within months
of their observation or faster (Mumma et al, 2009;
Giuranna et al, 2019; Webster et al., 2015; 2021).
If gas-dust interactions are important to the fast
destruction mechanism, the process of PBL growth
and collapse described in Moores et al. (2019) and
confirmed by Webster et al. (2021) would lead to an
amplification. This occurs because the frequency of
dust-particle interactions is directly related to the
atmospheric diffusivity – the same parameter driving
the bulk convection and mixing of the PBL.
As such, not only does daytime mixing dilute any
methane emitted near the surface, but such mixing
would also increase by orders of magnitude the reaction rate of any destruction mechanism. At night, not
only would emitted methane be trapped near the surface, but any gas-solid chemistry-based fast destruction mechanism would further be halted from acting,
revealing the true micro-seepage rate from the subsurface (Moores et al., 2019). If such a mechanism
exists for methane, it is likely to be important for
other species as well, providing a novel window into
atmospheric chemistry on Mars.
Acknowledgements: This work was funded in
part by the Natural Sciences and Engineering Research Council of Canada (NSERC) Discovery Grant
program. The authors also wish to thank the
NASA/Canadian Space Agency Mars Science Laboratory Participating Scientist Program.

References:
Atreya et al (2021) 43rd COSPAR Scientific Assembly. 43E1937A Abstract F3.3-0001-21 id. 1937.
Atreya et al (2006) Astrobiology, 6(3), 439–450.
Atreya et al (2011) Planetary and Space Science,
59(2‐3),
133–136.
https://doi.org/10.1016/j.pss.2010.10.008
Delory et al (2006) Astrobiology, 6(3), 451–462.
Giuranna et al. (2019) Nature Geoscience, 12(5),
326–332.
https://doi.org/10.1038/s41561‐019‐0331‐9
Jensen et al. (2014) Icarus, 236, 24–27.
https://doi.org/10.1016/j.icarus.2014.03.036
Korablev et al. (2021) Science Advances. 7:
eabe4386
Lefevre et Forget (2009) Nature. 460 720-723
Moores et al. (2007) Icarus. 192 (2) p. 417-433 doi:
10.1016/j.icarus.2007.07.003
Moores et al. (2019) Geophysical Research Letters,
46. https://doi.org/10.1029/2019GL083800
Mumma et al. (2009) Science, 323(5917), 1041–
1045. https://doi.org/10.1126/science.1165243
Newman et al. (2017) Icarus, 291, 203–231.
https://doi.org/10.1016/j.icarus.2016.12.016
Pathak et al. (2008) J. of Geophysical Research, 113,
E00A05. https://doi.org/10.1029/ 2007JE002967
Taylor et al. (2007) Boundary‐Layer Meteorology,
125(2),
305–328.
https://doi.org/10.1007/s10546‐007‐9158‐9
Thøgersen et al (2019) Icarus. 332 p. 14-18 doi:
10.1016/j.icarus.2019.06.025
Trainer et al. (2019) J. Geophys. Res. Planets. 124,
3000-3024. Doi: 10.1029/2019JE006175
Webster et al (2015) Science, 347(6220), 415–417.
https://doi.org/10.1126/science.1261713
Webster et al (2021) Astronomy and Astrophysics
650 A166 doi: 10.1051/004-6361/202040030