MARS ATMOSPHERE AS SEEN BY ATMOSPHERIC CHEMISTRY SUITE
ONBOARD EXOMARS TGO
O. Korablev, Space Research Institute of the Russian Academy of Sciences (IKI RAS), Moscow, Russia, F. Montmessin,
LATMOS/IPSL, UVSQ Université Paris-Saclay, Sorbonne Université, CNRS, Guyancourt, France, The ACS Team

Introduction: The Atmospheric Chemistry Suite
(ACS) instrument on board the ExoMars TGO dedicated atmospheric spacecraft has been successfully
operating since March 2018, delivering unique new
data. ACS includes three spectroscopic channels
covering from the near IR through thermal IR wavelengths. ACS NIR is a near-IR echelle spectrometer
with filtering of diffraction orders using an AOTF.
ACS MIR is a crossed dispersion echelle spectrometer, and ACS TIRVIM is a Fourier spectrometer [1].
MIR and NIR are used in solar occultation. These
channels remain in perfect health; their operation is
being continued. Compared to MIR, whose secondary dispersion grating has to be put in one selected
out of 11 positions, MIR delivers a homogeneous
and denser dataset. The cryogenic cooler of TIRVIM
stopped working in the end of 2019, approaching its
expected lifetime and short of one full MY. The
channel collected a corpus of atmospheric spectra,
mostly in nadir geometry.
The talk will introduce the ACS experiment and
outline its main findings, grouped in five sections,
methane, trace species, water and escape, CO, O2,
and CO2, the atmospheric structure.
Methane: The ACS/TGO spectra in the wavelength range where the strongest methane features
should be found, established a stringent upper limit
on CH4. The instrument registers faint absorption
lines of water vapour but not the neighbouring features of methane. First, ACS observations gave an
upper limit of 0.05 ppbv [2], while a more extensive
dataset covering MY34-35 resulted in the upper limit
of 0.02 ppbv or 20 ppptv [3]. The main difficulty
lies in contradiction between the "background" level
of methane by Curiosity [4], and the upper limit by
TGO. With a lifetime of 300 years [5], methane is
expected to add up, sooner or later exceeding the
TGO upper limit. Assuming Gale the only source
and only the background methane (0.41 ppbv), the
accumulation time after which TGO should detect
methane is 24y [2]. There are many places on Mars
whose geological structure makes them potential
sources of methane; the estimate does not take into
account spikes and the improved TGO upper limit.
Including these factors makes the accumulation time
even shorter. Mesoscale simulations of methane
dissipating from Gale under the TGO constraints
suggest that not only it is the single source on the
planet, but a particular location of the putative source
within the crater rim can solely explain the observed
background and spikes [6, 7]. Moores et al. [8] suggested that the height of the planetary boundary layer
in the crater decreases at night, limiting mixing and

leading to the accumulation of methane. Indeed,
recent SAM-TLS daytime measurements are only
upper limits of ≤0.2 ppbv [6]. The night-time scenario could increase the accumulation time of methane or the allowable area on Mars where it can seep.
Still, to comply with the TGO upper limits, an as yet
unknown mechanism for the rapid methane destruction is needed. ACS continues regular CH4 measurements.
Trace gases and faint absorptions: "In place of
methane", i.e. in the spectral range containing the
strongest methane spectral features, two new absorbers are discovered in the ACS MIR spectra. In a
range of 2900–3300 cm–1, 30 observed lines were
absent from spectroscopic databases. Their frequencies matched the theoretically calculated P-, Q-, and
R-branches of the magnetic dipole or electroquadrupole band of the main CO2 isotopologue. This
band, forbidden for electro-dipole absorption, has
been never observed or numerically calculated. The
relative depth of branches suggested its magnetodipole nature [9]. Later, the even weaker electric
quadrupole system was also found in the ACS spectra [10].
In the same spectral range, we detected a weak
infrared ozone band overlapping one of methane's Pbranch features. This band has never been seen in the
Mars atmosphere before. Observations in the IR
make it possible to reconstruct vertical profiles and
explore the lowermost layer of ozone [11, 12].
Hydrogen chloride (HCl) was confidently identified by 12 lines in the ACS MIR spectra [13, 14]. As
confirmed by NOMAD, the HCl mixing ratio
reaches 4 ppbv, significantly higher than previous
upper limits. The gas appeared after the MY34 GDS
and in perihelion of MY35, while outside the dusty
season it remained ≤0.1 ppbv. A possible mechanism
of HCl formation involves atmospheric dust (see talk
by K Olsen et al.) The high fidelity of the collected
spectra makes it possible to estimate the ratio
H37Cl/H35Cl [15]. Unlike for other volatiles on Mars,
the isotopic anomaly is indistinguishable. Given its
short lifetime, fractionation mechanisms in HCl are
too weak to be noticed.
A 1-D chemical model in the meteorological context from LMD MGCM [16] reproduces the seasonal
and vertical behaviour of HCl. It also suggests that
chlorine chemistry might shorten the lifetime of
methane down to weeks or days, potentially resolving the methane conundrum. This hypothesis needs
further validation.
The sensitive search for more trace species resulted in improving of several upper limits. ACS has

established upper limits of ≤50 pptv on C2H6, ≤1
ppbv on CH3Cl, ≤0.4 ppbv on OCS [17], ≤0.1 ppbv
on phosphine PH3 [18]. The model-predicted NO2
abundance of ~0.2 ppmv is tentatively confirmed
(see talk by A. Trokhimovskiy et al.)

properties were investigated [30, 31] (See talk by
Stcherbinine et al.)
The effort to constrain the abundances of oddhydrogen species OH and HO2 from MIR spectra is
ongoing (See talk by J. Alday et al.)

Water and escape: ACS measures water vapour
profiles in occultation with NIR (1.38-µm band) and
with MIR in broader range, covering isotopologues
and the strongest 2.56-µm band. The vertical distribution of water vapour was measured up to 120 km
[19-21]. During the MY34 GDS, smaller scale
events like the MY34 C-storm, and the perihelion
season of MY35 the H2O mixing ratio of ~100 ppmv
is observed at 100 km and higher. The Lyman-alpha
measurements by MAVEN showed simultaneous
increase of escaping H atoms during the MY34 Cstorm, implying such dust events can boost the planetary H loss by a factor of 5–10 [22]. Still, ACS measurements and modelling suggest that the seasonal
signal is water elevation and H escape is comparable
to that during GDS [21, 23]. Given the duration of
the perihelion dusty season and the fact that GDS
occurs every third MY at most, we infer the seasonal
contribution to escape is prevailing [23].
ACS MIR measures altitude profiles of H and O
isotope ratios in H2O. HDO and H2O profiles combined with expected photolysis rates, also reveal the
prevalence of the perihelion season for the formation
of atomic H and D at altitudes relevant for escape
[20]. This study shows that while condensationinduced fractionation is the main driver of variations
of D/H in water vapour, the differential photolysis of
HDO and H2O mainly controls the isotopic composition of the dissociation products. Observed variations
of the D/H ratio, lower HDO/H2O ratios in the colder
regions of Mars, the impact of the MY34 GDS were
simulated with GCM [24-26] (See talk by Vals et al.)
MIR measurements allowed, for the first time, the
measurement of vertical profiles of the 18O/16O and
17
O/16O ratios in atmospheric water vapour. They are
enriched with respect to Earth (δ18O = 200 ± 80‰
and δ17O = 230 ± 110‰ and do not show any evidence of altitudinal variations [27].
Both NIR and MIR ACS water profiles are accompanied with simultaneous density-temperature
profiles retrieved from CO2 absorptions, allowing to
assess the saturation state. Fedorova et al. [19, 28]
concluded about nearly ubiquitous supersaturation of
water, particularly during the dust season, thereby
promoting water escape on an annual average. The
robustness of the supersaturation detected in the
specific terminator geometry was supported by
hourly-resolved OU MGCM modelling with assimilation [29]. Within the same framework, ACS water
profiles serve to define the varying hygropause altitude (See talk by J. Holmes et al.)
Water ice clouds detected by ACS MIR and
TIRVIM were studied in the context of atmospheric
and water profiles [19]. The seasonal distribution of
condensation clouds and also dust, and the aerosol

CO, O2 and CO2: The first vertical profiles of
carbon monoxide CO before and during the MY34
GDS were retrieved from MIR spectra at 2.35 µm
[32]. They showed a prominent depletion of CO up
to 100 km, pointing to the importance of CO oxidation during wetter GDS conditions. A complete climatology of CO vertical profiles from the beginning
of the mission to the end of MY35 was obtained
from NIR data using a weaker CO band at 1.6 µm,
overlapped with stronger CO2 band. CO profiles
show strong enrichment of 3000-4000 ppmv at 10–
20 km in southern winter and early spring in midhigh southern latitudes. An enrichment is also found
above 50 km at equinoxes [33]. The mean CO VMR
is ~950 ppmv, higher than Curiosity in situ measurements [34] and closer to CRISM, PFS and
NOMAD nadir results [35–37].
Another incondensable species, O2, is also measured by ACS NIR in its 0.76 µm band. The as yet
unpublished two-MY dataset includes O2 profiles
measured up to 40-50 km and exhibit enrichments
similar to those observed in CO. The average O2
mixing ratio is 1800 ppmv; the O2/CO ratio of ~2.
(see talk by A. Fedorova et al.)
Isotopic ratios in atmospheric CO2 were monitored using MIR data to obtain the O and C isotopic
composition at 70–130 km. Their vertical trends are
consistent with the expectations from diffusive separation above the homopause, with average values
below showing Earth-like fractionation. Results
suggest that at least 20-40% of primordial carbon has
been lost from Mars throughout history [38].
Atmospheric structure: ACS TIRVIM measures spectra of the outgoing thermal radiation in
nadir. The spectral range (600–1300 cm–1) includes
the 15-µm CO2 band. Inversion of spectra in this
band yields vertical temperature profiles from the
surface up to 60 km of altitude. The continuum spectrum at shorter wavelength permit to measure surface
temperatures and estimate column optical depths of
dust and H2O clouds. TIRVIM operated from
Ls=143° MY34 to Ls=115° MY35. The nadir coverage during this period varied within 500–9000 spectra per orbit, allowing, despite several gaps due to
hardware issues, a uniquely dense coverage of the
diurnal cycle. Two methods of inversion are developed and validated against MCS/MRO profiles [39,
40]. Most differences between TIRVIM and MCS
atmospheric temperatures can be attributed to differences in vertical sensitivity.
The atmospheric thermal structure and the column dust content during the MY34 GDS was analysed, capturing its evolution and the atmospheric
response to the changing dust load. The global storm

caused asymmetric atmosphere heating, predominantly in the southern hemisphere, and changed
diurnal contrast of atmospheric thermal structure.
Also, at the peak of the GDS we see a reduced diurnal contrast of surface temperatures [40]. To understand the meteorology of GDS, in particular the
atmospheric properties that cannot be measured
directly, such as wind, data assimilation was applied
using LMD MGCM and the Local Ensemble Transform Kalman Filter [41]. The model predicts the
atmospheric state every three hours. At the peak of
the storm, a strong asymmetry developed in the
midlatitude jets, modifying the diurnal and semidiurnal tides. The reconstructed surface pressure was
verified against Curiosity measurements (See talk by
R. Young et al.)
Thermal tides in the Martian atmosphere were
analysed using temperature the TIRVIM nadir profiles near the northern summer solstice of MY35.
The observations have a full local time coverage,
which enables analyses of daily temperature anomalies. Wave mode decomposition showed dominant
diurnal tide, important semi-diurnal tide and diurnal
Kelvin wave, with maximal amplitudes of 5, 3, and
2.5 K, respectively, at 10–100 Pa. The results agree
with the LMD MGCM, yet with noticeable earlier
phases of diurnal and semi-diurnal tides [41].
The thermal structure at the limb was reconstructed from ACS MIR occultation spectra in CO2
bands. Belyaev et al. [42] using the strongest 4.3-µm
band of CO2 retrieved the atmospheric structure up
to 180 km and traced the homopause and mesopause
over seasons. The homopause was detected from 80
km at aphelion to 110 km at perihelion thorough
MY34-35, depending on dust activity. The
mesopause altitude rises from 70-90 km in the highwinter latitudes to 130-150 km in summer (See talk
by D. Belyaev et al.)
Starichenko et al. have used MIR profiles to
study the gravity waves at 20-160 km. For the first
time such a wide altitude range was analysed. The
waves’ amplitude increases up to ~100 km, and they
break up above, in the mesopause region 100–130
km (See talk by E. Starichenko et al.)
.
References
[1] Korablev, O. et al. Space Sci. Rev. 214, 7
(2018).
[2] Korablev, O. et al. Nature 568, 517 (2019).
[3] Montmessin, F. et al. Astron. Astrophys. 650,
A140 (2021).
[4] Webster, C. et al. Science 360, 6393, 10931096 (2018).
[5] Lefevre, F., Forget, F. Nature 460, 720
(2009).
[6] Webster, C. R. et al. A&A 650, A166 (2021).
[7] Luo, Y., et al. Earth Space Sci. 8, e01915
(2021).
[8] Moores, J. E. et al. Geophys. Res. Lett. 46,

9430 (2019).
[9] Trokhimovskiy, A. et al. A&A 639, A142
(2020)
[10] Yachmenev, A. et al. J. Chemical Phys. 154,
211104 (2021).
[11] Olsen, K. S. et al. A&A 639, A141 (2020).
[12] Olsen, K. S. et al. Seasonal changes in the
vertical structure of ozone in the Martian lower atmosphere and its relationship to water vapor. J. Geophys. Res. Planets, Submitted.
[13] Korablev, O. et al. Sci. Adv. 7, eabe4386
(2021).
[14] Olsen, K. S. et al. A&A 647, A161 (2021).
[15] Trokhimovskiy, A. et al. A&A 651, A32.
(2021).
[16] Taysum, B., et al. Martian atmospheric
chemistry of HCl: implications for the lifetime of
atmospheric methane. J. Geophys. Res. Planets,
Submitted.
[17] Braude, A. S. et al. Astron. Astrophys. 658,
A86 (2022).
[18] Olsen, K. S. et al. A&A 649, L1 (2021).
[19] Fedorova, A. A. et al. Science 367, 297
(2020).
[20] Alday, J. et al. Nat. Astron. 5, 943 (2021).
[21] Belyaev, D. A. et al. Geophys. Res. Lett. 48,
e93411 (2021).
[22] Chaffin, M. S. et al. Nat. Astron. 5, 1036
(2021).
[23] Montmessin, F. et al. Reappraising the production and transfer of hydrogen atoms from the
middle to the upper atmosphere of Mars at times of
elevated water vapor. J. Geophys. Res. Planets,
Submitted.
[24] Rossi, L. et al. Geophys. Res. Lett. 48,
e90962 (2021).
[25] Vals, M. et al. Improved modeling of Mars'
HDO cycle using a Mars' Global Climate Model. J.
Geophys. Res. Planets, Submitted.
[26] Rossi, L. et al. The HDO cycle on Mars:
comparison of ACS observations with GCM simulations. J. Geophys. Res. Planets, Submitted.
[27] Alday, J. et al. A&A 630, A91 (2019).
[28] Fedorova, A. A. et al. A two-Martian year
survey of the water vapor saturation state on Mars. J.
Geophys. Res. Planets, Submitted.
[29] Holmes, J. A., et al. Global variations in water vapor and saturation state throughout the Mars
Year 34 dusty season. J. Geophys. Res. Planets,
Submitted.
[30] Luginin, M. et al. J. Geophys. Res. Planets
125, e06419 (2020).
[31] Stcherbinine, A. et al. J. Geophys. Res.
Planets 125, e06300 (2020).
[32] Olsen, K. S. et al. Nat. Geosci. 14, 67
(2021).
[33] Fedorova, A. A. et al. Climatology of the
CO vertical distribution on Mars based on ACS TGO
measurements. J. Geophys. Res. Planets, Submitted.
[34] Trainer et al. J. Geophys. Res. Planets 124,

3000, (2019).
[35] Smith et al. Icarus, 301, 117 (2018).
[36] Bouche et al. J. Geophys. Res. Planets
126(2), e2020JE006480 (2021).
[37] Smith et al. Icarus, 362, 114404 (2021).
[38] Alday, J., et al. J. Geophys. Res. Planets
126, e06992 (2021).
[39] Guerlet, S. J., et al. Geophys. Res. Planets
127, e07062 (2022).
[40] Vlasov, P., et al. Martian atmospheric thermal structure and dust distribution during the MY 34
global dust storm from ACS TIRVIM nadir observations. J. Geophys. Res. Planets, Submitted.
[41] Young, R., et al. Assimilation of temperatures and column dust opacities measured by
ExoMars TGO-ACS-TIRVIM during the MY34
Global Dust Storm. J. Geophys. Res. Planets, Submitted.
[42] Fan, S., et al. Geophys. Res. Lett. 49,
e2021GL097130 (2022).
[43] Belyaev, D., et al. Thermal Structure of the
Middle and Upper Atmosphere of Mars from
ACS/TGO CO2 Spectroscopy. J. Geophys. Res.
Planets, Submitted.
[44] Starichenko, E., et al. J. Geophys. Res. Planets 126, e06899 (2021).