IMPACTS OF HETEROGENEOUS CHEMISTRY ON VERTICAL PROFILES
OF MARTIAN OZONE.
M. A. J. Brown, School of Physical Sciences, The Open University, Milton Keynes, U.K. (megan.brown@open.ac.uk),
M. R. Patel, S. R. Lewis, J. A. Holmes, G. J. Sellers, P. M. Streeter, A. Bennaceaur, The Open University, Milton
Keynes, U.K., G. L. Villanueva, Planetary Systems Laboratory, NASA Goddard Space Flight Center, Greenbelt, MD,
USA, G. Liuzzi, Planetary Systems Laboratory, NASA Goddard Space Flight Center, Greenbelt, MD, USA, Department of Physics, American University, Washington, DC, USA, A. C. Vandaele, Royal Belgian Institute for Space
Aeronomy (BIRA-IASB), Brussels, Belgium.

Background
Ozone (O3 ) was first detected in the martian atmosphere
in 1972 by Mariner 7 and again by Mariner 10 (Barth
et al., 1973). Due to its chemical sensitivity and relatively short dayside lifetime (2–3 hours), it is often used
as a proxy for other, highly reactive species, which cannot be measured directly with remote sensing (Clancy
and Nair, 1996). Such chemical species include hydroxyl radicals (H, OH, and HO2 , collectively known
as HOx ), which react with ozone, and participate in
its destruction. The two primary destructive pathways
of ozone are from UV light (via photolysis) and reactions with HOx species (Clancy and Nair, 1996; Lefèvre
et al., 2004). The latter can cause a set of chain reactions which create more HOx species and lead, in turn,
to further ozone destruction. In addition to ozone destruction, HOx also catalyse the reaction between carbon
monoxide (CO) and molecular oxygen (O), recombining
to form carbon dioxide (CO2 ), the primary constituent
of the martian atmosphere. The underlying chemistry
of HOx is thus crucial for understanding the stability of
the martian atmosphere, as well as its composition.
HOx are formed by the photolysis of water vapour,
resulting in the well known photochemical anti-correlation
between ozone and water vapour. This relationship has
been used to understand water vapour transport and seasonal variation (Montmessin and Lefèvre, 2013).
Ozone is underpredicted in martian climate models,
implying there are some chemical reactions which are
missing or incorrect in models. Heterogeneous reactions of HOx on the surface of water ice have previously
been offered as an explanation for the underprediction
of ozone in climate models (Lefèvre et al., 2008, 2021).
However, in previous work, including this chemistry
has led to an overprediction of ozone (Clancy et al.,
2016). In the adsorption of HOx species, water ice acts
as a sink, and the HOx abundance decreases, allowing
the ozone abundance to increase. As a result, the relationship between ozone and water ice has been used
as a proxy for the presence of heterogeneous reactions
(Clancy et al., 2016). Previous studies with heterogeneous chemistry have only investigated total column
abundances of species, which are biased towards low
altitude ozone. One way to reduce this bias is to use

vertical profile observations with a unit of mass mixing
ratio, giving equal weighting across all altitudes.
We investigate reactions between highly reactive
HOx and water ice as a potential explanation for the
modelled underprediction of ozone by using a 1-D model
and observed vertical profiles from the NOMAD instrument, aboard ExoMars TGO. The vertical relationship
between water ice and ozone is used as a proxy for heterogeneous reactions, using profiles from Patel et al.
(2021) and Liuzzi et al. (2020). We then use a 1-D
model to simulate two water vapour scenarios in order
to analyse the ozone – water ice relationship under different circumstances with heterogeneous chemistry. By
using a 1-D model, the difference in ozone between the
heterogeneous and gas-phase-only chemical simulation
can be isolated as purely chemical, without the addition
of horizontal transport.

Vertical Cross-Correlation
Vertical profiles are divided into two groups: high northern and high southern latitudes (> 45◦ ) from LS =
0◦ − 180◦ . The two regions have contrasting water
vapour abundances during this time of year, with the
southern and northern latitudes having a low and high
vapour abundance respectively. We conduct a vertical cross-correlation between observed profiles of ozone
and water ice from the NOMAD instrument to identify
any relationship within the two regions.
The cross-correlation technique is often used in terrestrial studies between two time series, in order to determine if there is a correlated time lag between the
variables. It also determines the strength of the correlation, as well as the length of the time displacement lag
(e.g. Arattano and Marchi (2005); Peppa et al. (2017)).
A cross-correlation is often used when one variable is
expected to influence the other, with some delay in the
response. In this study, altitude (rather than time) is displaced in order to identify any vertical patterns between
the ozone and water ice, which, due to the adsorption
of HOx in water-ice clouds, may impact ozone across
different altitudes. If there is any variation of ozone in
the water-ice cloud, this may be detected by the vertical
cross-correlation within a few kilometres displacement.

Figure 1: A cross-correlation on a pair of profiles: (left panel)
vertical profiles of (blue) water ice and (red) ozone; crosses are
datapoints and lines are the interpolated profiles every 2 km;
(right panel) the (green) vertical correlation between the profiles across all displacements. A positive displacement of 4 km
corresponds to a shifting of the water ice profile up by 4 km.
The pink area marks the correlations which are not significant according to a Student’s T-test at 5% significance. The
red cross indicates the most significant vertical correlation
of the cross-correlation. Profiles are from NOMAD/TGO at
LS = 187◦ , latitude 70.5◦ S, 2020.04.21 11:04:57 UTC.

Figure 1 shows an example cross-correlation between a pair of vertical profiles. The left panel shows the
(red) ozone and (blue) water ice vertical profiles, while
the right panel shows the correlation of the profiles at
different altitude displacements. The highest correlation
occurs at -10 km displacement, when the peak in water
ice is 10 km below the peak in ozone. This analysis
was conducted on all water ice and ozone pairs for both
latitudinal regions, which met certain minimum conditions. Such conditions consisted of minimum values and
a minimum number of datapoints within each profile.

1-D Model
The 1-D model used in this work is derived from the
Open University modelling group Mars GCM (MGCM),
which exists as a collaboration between the Laboratoire de Météorologie Dynamique (LMD), the Open
University, the University of Oxford, and Instituto de
Astrofı́sica de Andalućia (Forget et al., 1999).
The heterogeneous chemical scheme has been modified from the ASIS scheme from Cariolle et al. (2017),
used in Lefèvre et al. (2021), to better represent the
adsorption of HOx as a sink. This is done by adding
chemical species which represent the adsorbed status of
HOx , as given by the following;
OH + ice → iceOH ,

where iceOH is an example of the added species.

(1)

Figure 2: Averaged profiles (212 vertical profiles) > 45◦ S
between LS = 100◦ − 120◦ of (green line) observed water
ice, and (red bars) the difference in ozone (observed – modelled). Positive values indicate an underprediction of ozone.
Modelled data are taken from the MGCM, assimilated with
temperature and dust retrievals from Mars Climate Sounder.

Two scenarios with a high and low water vapour
abundance are simulated with the 1-D model to represent
high northern and southern latitudes respectively. The
scenarios are a simplified representation of high latitude
regions and during the aphelion season. To reduce the
parameter differences between the high and low water
vapour scenarios, the latitudes and local times are kept
consistent at latitude 0◦ and 1200 local solar time (LST);
only the time of year is changed, with the low and high
water vapour scenario at LS = 60◦ and LS = 180◦ respectively. The solar insolation is lower in the LS = 60◦
simulation, resulting in a lower water vapour abundance
due to lower temperatures.

Results
Ozone is currently underpredicted by martian GCMs in
both total column abundance and vertical profiles (e.g.
Daerden et al. (2019); Lefèvre et al. (2008); Patel et al.
(2021)). Comparing vertical profiles observations with
simulations, the ozone deficit coincides with increases
in water ice abundance, as seen in Figure 2.
From the cross-correlation, we find a positive vertical correlation between ozone and water ice observations
at high southern latitudes when water vapour abundance
is low. The left panel of Figure 3 shows a histogram
of the significant correlations from each set of crosscorrelations of ozone and water ice profile pairs. The
histogram shows the altitude displacement, binned ev-

REFERENCES

Figure 3: Histogram of altitude displacements for significant, positive, vertical correlations between ozone and water
ice observed profiles for (left) southern latitudes (> 45◦ S)
LS = 0◦ − 180◦ and (right) northern latitudes (> 45◦ N) between LS = 30◦ − 180◦ . α shows the significance level, and
r ≥ 0 indicates positive correlations only. Bins cover 5 km,
with the y axis label denoting the centre of the bin (e.g. the
0 km bin includes values from 2.5 km to 2.5 km).

ery 5 km. The frequency density peaks when the altitude
displacement is between -2.5 − 2.5 km, implying that a
large proportion of ozone and water ice profiles have a
positive correlation when they are within 5 km of each
other.
In contrast, there is no such clear peak in the histogram for the high northern latitudes (right panel, Figure 3). The altitude displacements are spread across
multiple bins without any distinguishing peaks. This
suggests that there is no clear vertical relationship between ozone and water ice. At the high northern latitudes, the water vapour abundance is high, which may
impact the effect of heterogeneous chemistry on ozone.
A key difference between the two regions is the water
vapour abundance; northern latitudes have a high water
vapour abundance, while the southern latitudes have a
low vapour abundance.
We replicate these conditions in a 1-D model with
simplified scenarios of high and low water vapour abundance. By comparing the difference in ozone with a
heterogeneous and gas-phase-only model, we can isolate the effects of heterogeneous chemistry on ozone.
There is a large ozone difference between the heterogeneous and gas-phase-only run for the low abundance
simulation, occurring when water-ice clouds form. At
these altitudes water ice is present, the HOx abundance
decreases significantly, and, as a result, ozone abundance
increases, as the ozone destruction rate decreases.

On the other hand, the ozone difference in the high
vapour scenario is minimal. The HOx abundance is
much greater and, despite the water ice abundance being three times greater than in the low vapour scenario,
the reduction in HOx has little effect on ozone. This
is due to the high HOx abundance, formed by water
vapour photolysis. The heterogeneous reactions remove
a smaller relative amount of HOx , resulting in a high
HOx abundance still available to destroy ozone. The adsorption of HOx onto water ice under such circumstances
is dominated by the high HOx abundance, and there is
little difference in ozone between the heterogeneous and
gas-phase-only simulation.

Conclusions
By using a combination of vertical profile observations
and 1-D modelling, we find that the water vapour abundance affects the impact that heterogeneous reactions
have on ozone.
When water vapour abundance is low, HOx is adsorbed at altitudes water ice is present, lowering the
HOx abundance, and thus allowing ozone to form. This
is shown in the 1-D model where there is a large ozone
difference in favour of the heterogeneous simulation in
the low vapour scenario. Similarly in the vertical profile observations, there is a strong, positive correlation
between ozone and water ice at high southern latitudes
when vapour is low. Using this relationship as a proxy,
the results from the vertical cross-correlation could indicate heterogeneous chemistry.
In areas of high water vapour abundance, the amount
of HOx adsorbed relative to the total HOx abundance
is minimal, and there is little difference in ozone between the heterogeneous and gas-phase-only simulation.
Therefore, it can be expected that the effect of heterogeneous chemistry cannot be observed through the variation in ozone abundance. The positive relationship between ozone and water ice, which is used as a proxy for
heterogeneous chemistry, is not expected to be present
under circumstances with high water vapour abundance.
The lack of a clear vertical correlation between ozone
and water ice at high northern latitudes can be explained
by the high water vapour abundance (and hence high
HOx abundance).
The inclusion of heterogeneous chemistry could be
used to explain at least some of the ozone deficit seen in
GCMs, especially when water vapour abundance is low.

References
C. A. Barth, C. W. Hord, A. I. Stewart, A. L. Lane, M. L. Dick,
and G. P. Anderson, “Mariner 9 ultraviolet spectrometer experiment: Seasonal variation of ozone on mars,” Science,
vol. 179, no. 4075, pp. 795–796, 1973.

REFERENCES
R. T. Clancy and H. Nair, “Annual (perihelion-aphelion) cycles in the photochemical behavior of the global mars atmosphere,” Journal of Geophysical Research: Planets,
vol. 101, no. E5, pp. 12785–12790, 1996.
F. Lefèvre, S. Lebonnois, F. Montmessin, and F. Forget,
“Three-dimensional modeling of ozone on mars,” Journal
of Geophysical Research: Planets, vol. 109, no. E7, 2004.
F. Montmessin and F. Lefèvre, “Transport-driven formation
of a polar ozone layer on mars,” Nature Geoscience, vol. 6,
p. 930, 2013.
F. Lefèvre, J.-L. Bertaux, R. T. Clancy, T. Encrenaz, K. Fast,
F. Forget, S. Lebonnois, F. Montmessin, and S. Perrier,
“Heterogeneous chemistry in the atmosphere of mars,” Nature, vol. 454, p. 971, 2008.
F. Lefèvre, A. Trokhimovskiy, A. Fedorova, L. Baggio, G. Lacombe, A. Määttänen, J.-L. Bertaux, F. Forget, E. Millour,
O. Venot, Y. Bénilan, O. Korablev, and F. Montmessin,
“Relationship between the ozone and water vapor columns
on mars as observed by spicam and calculated by a global
climate model,” Journal of Geophysical Research: Planets,
vol. 126, no. 4, p. e2021JE006838, 2021. e2021JE006838
2021JE006838.
R. T. Clancy, M. J. Wolff, F. Lefèvre, B. A. Cantor, M. C.
Malin, and M. D. Smith, “Daily global mapping of mars
ozone column abundances with marci uv band imaging,”
Icarus, vol. 266, pp. 112–133, 2016.
M. R. Patel, G. Sellers, J. P. Mason, J. A. Holmes, M. A. J.
Brown, S. R. Lewis, K. Rajendran, P. M. Streeter, C. Marriner, B. G. Hathi, D. J. Slade, M. R. Leese, M. J. Wolff,
A. S. J. Khayat, M. D. Smith, S. Aoki, A. Piccialli,
A. C. Vandaele, S. Robert, F. Daerden, I. R. Thomas,
B. Ristic, Y. Willame, C. Depiesse, G. Bellucci, and J.-J.
Lopez-Moreno, “Exomars tgo/nomad-uvis vertical profiles
of ozone: 1. seasonal variation and comparison to water,”
Journal of Geophysical Research: Planets, vol. 126, no. 11,
p. e2021JE006837, 2021. e2021JE006837 2021JE006837.
G. Liuzzi, G. L. Villanueva, M. M. Crismani, M. D. Smith,
M. J. Mumma, F. Daerden, S. Aoki, A. C. Vandaele, R. T.
Clancy, J. Erwin, I. Thomas, B. Ristic, J.-J. Lopez-Moreno,
G. Bellucci, and M. R. Patel, “Strong variability of martian water ice clouds during dust storms revealed from exomars trace gas orbiter/nomad,” Journal of Geophysical Research: Planets, vol. 125, no. 4, p. e2019JE006250, 2020.
e2019JE006250 2019JE006250.
M. Arattano and L. Marchi, “Measurements of debris flow
velocity through cross-correlation of instrumentation data,”
Natural Hazards and Earth System Sciences, vol. 5, no. 1,
pp. 137–142, 2005.
M. V. Peppa, J. P. Mills, P. Moore, P. E. Miller, and J. E. Chambers, “Brief communication: Landslide motion from cross

correlation of uav-derived morphological attributes,” Natural Hazards and Earth System Sciences, vol. 17, no. 12,
pp. 2143–2150, 2017.
F. Forget, F. Hourdin, R. Fournier, C. Hourdin, O. Talagrand,
M. Collins, S. R. Lewis, P. L. Read, and J.-P. Huot, “Improved general circulation models of the martian atmosphere from the surface to above 80 km,” Journal of Geophysical Research: Planets, vol. 104, no. E10, pp. 24155–
24175, 1999.
D. Cariolle, P. Moinat, H. Teyssèdre, L. Giraud, B. Josse, and
F. Lefèvre, “Asis v1.0: an adaptive solver for the simulation of atmospheric chemistry,” Geoscientific Model Development, vol. 10, no. 4, pp. 1467–1485, 2017.
F. Daerden, L. Neary, S. Viscardy, A. G. Muñoz, R. Clancy,
M. Smith, T. Encrenaz, and A. Fedorova, “Mars atmospheric chemistry simulations with the gem-mars general
circulation model,” Icarus, vol. 326, pp. 197 – 224, 2019.