MARS ATMOSPHERIC THERMAL STRUCTURE DURING THE 1ST YEAR
OF TGO SOLAR OCCULTATION MEASUREMENTS.
Miguel Angel López-Valverde, (valverde@iaa.es), B. Funke, A. Brines, A. Modak, A. Stolzenbach, F.
González-Galindo, B. Hill, J. J. López-Moreno, Instituto de Astrofı́sica de Andalucia (IAA/CSIC), Granada, Spain
(valverde@iaa.es), I. Thomas, L. Trompet, J. Erwin, Belgian Royal Institute for Space Aeronomy, Brussels, Belgium,
S. Aoki, University of Tokyo, Japan, Gerónimo Villanueva, NASA Goddard Space Flight Center, USA, Giuliano Liuzzi, American University, Washington DC, USA, F. Forget, Laboratoire de Meteorologie Dynamique, Paris, France,
Giancarlo Bellucci, Institute for Space Astrophysics and Planetology, Italy, Manish Patel, Open University, Milton
Keynes, UK, Bojan Ristic, Frank Daerden, Ann-Carine Vandaele, Belgian Royal Institute for Space Aeronomy,
Brussels, Belgium.

Introduction
The detailed variation of temperature and density with
altitude is of paramount importance to characterize the
atmospheric state and to constrain the chemistry, radiation and dynamics as a whole. Only with an accurate description of vertical variations with the best
possible spatial and resolution we can aim at describing
the key exchanges between atmospheric layers. These
are needed for transport of radiatively active species,
for energy budget modeling, including photochemistry,
gravity wave propagation and global dynamics, and for
atmospheric escape processes, to mention a few.
The power of solar occultation The NOMAD and
the Atmospheric Chemistry Suite (ACS) are the two key
instruments for atmospheric research on board the ExoMars 2016 Trace Gas Orbiter, primarily devoted to trace
gas detection and mapping [15, 4]. The solar occultation observational strategy of these two instruments
offers a unique opportunity to sound also up to high altitudes with an unprecedented vertical resolution in order
to perform unprecedented sounding at mesospheric and
thermospheric altitudes [6].
Goals This is the long term target of our team, and in
this work we present first results. We analyzed transmittance spectra obtained from the NOMAD solar occultation channel, with a state-of-the-art retrieval scheme,
adapted from Earth to Mars conditions and geometry,
to derive temperature and density up to about 100 km.
We will present the quality and robustness of the inversions and its application to the study of the variability
of the Martian thermal structure in the troposphere and
mesosphere during the 1st first year of the TGO science
phase, from April 2018 to March 2019, which covers approximately two Mars seasons. These range from solar
longitudes Ls 160◦ to Ls 360◦ , i.e., the Southern spring
and summer seasons, sometimes collectively named as
the Martian perihelion period.
The results presented here follow upon a recent work
submitted to a special issue devoted to the analysis of
the first year of TGO observations [7].

Figure 1:
Example of our cleaning method applied to diffraction order 149 in one specific scan
(20180423 204351 1p0a SO A I). Top panel, all the
original calibrated transmittance spectra taken in this scan,
each tangent altitude with a different color, showing clear
bending effects and spectral shifts. Bottom panel, the same
spectra after cleaning, which is only applied to a subset of
the full diffraction order on purpose. Vertical red solid lines
indicate the CO2 spectral lines positions (Hitran2016).

Datasets
The NOMAD spectrograph operationally measures the
atmospheric absorption in several diffraction orders. The
data analyzed here are calibrated transmittances in the
so called order 149, containing moderately strong CO2
lines. The nominal measurement uncertainties seem to
contain small but clear systematics which are under investigation. We use here an estimation of their random
component, as a first approach. Other effects which
are corrected for during a “pre-processing phase” of our
data analysis include spectral shifts and a residual spec-

tral bending across each diffraction order, which vary
significantly from scan to scan, as they depend on temperature effects on the detector, on TGO orbital conditions which in turn impact the space signals, and on
drifts through time in the behavior of the grating’s blaze
response [5, 13]. These effects and their correction are
shown for one specific scan in Figure 1.
Let us note in particular the importance of a correct
handling of the “bending” mentioned above for a proper
description of the baseline continuum, specially at high
altitudes, where a deficient correction may introduce biases larger than the actual absorption lines. The bending
spectral shape can approximately be characterized by a
polynomial function but its precise removal requires to
take into account the gas absorption lines, i.e., a detailed
line-by-line forward model calculation.

Retrieval Scheme
Our retrieval scheme is based on a state-of-the-art lineby-line forward model called KOPRA ([12]) and an inversion control program (RCP, [16]) well tested in Earth
remote sounding experiments. RCP performs a global
fit of 1-D vertical profiles of radiance spectra, assuming
homogeneity in the horizontal along the line-of-sight.
RCP solves iteratively the inverse problem (Rodgers,
2000) until convergence is achieved. The calculation
of the model spectra and Jacobians is performed with

Figure 2:

Retrieved temperature profile from scan

20180423 204351 1p0a SO A I with error bars (topleft panel), where the red dashed line is the a priori profiles,
and the green dashed line is the CO2 condensation temperature, added for reference. Bottom-left: rows of the Averaging
Kernel matrix selected at some altitudes. Top-right: retrieval
error in the Temperature profile. Bottom-right: Vertical
resolution, as given by the widths of the averaging kernel
rows.

KOPRA, in each step of the iteration, including a hydrostatic adjustment in each iteration, to account for the
varying atmospheric state when the thermal structure
is changed, and therefore to guarantee realistic results
in the temperature/pressure structure. The hydrostatic
adjustment assumes a reference pressure at the lowermost tangent altitude available, and this value is taken
from the a priori reference atmosphere. The a priori
temperature and density profiles are taken from a specific run of the Mars Global Climate Model developed
at the Laboratoire de Météorologie Dynamique (LMDMGCM) [2], using the recent implementations of the
water cycles [10] and the dust scenarios appropriate for
MY34 [8].
Figure 2 shows the performance of the temperature
inversion in one specific scan obtained on June 20th,
2018, for Ls 196◦ and at 55◦ N latitude, where a heavy
dust loading was found below about 20 km. In addition
to the noise propagation, the vertical resolution, as given
by the width of the Averaging Kernels rows, is essential
for comparing results to other retrievals and to models.
With our optimal regularization, the vertical resolution
varies with the tangent altitude, between 3 km and 9 km,
and typical retrieval errors vary between 1.5-3.5 K.

Retrieved atmospheric temperatures
We describe the over 330 retrieved profiles during MY34
in detail in [7]. Here we highlight a few results which
will be discussed in the Workshop.
Figure 3 shows all the temperature profiles retrieved
in this work, put together in a single panel, with a similar plot for the LMD GCM profiles (our a priori). The
dashed lines in blue and orange represent the envelope
of all possible temperatures anywhere and anytime on
Mars, taken from [11], who compiled data from instruments Mars Climate Sounder and Thermal Emission Spectrometer, as well as from the Mars Climate
Database simulations for extreme conditions. An overall
agreement exists in the lower atmosphere but at mesospheric altitude the NOMAD temperatures are globally
colder than in the GCM by about 10 K.
The atmospheric thermal structure during the two
Mars seasons in MY34 (the Martian perihelion half-year
period) and shown in Figure 4 is characterized by the
strong MY34 global dust storm (GDS) episode, a well
documented event peaking around Ls 195◦ -210◦ [3, 8].
Our results show that the GDS warmed the atmosphere
significantly at all altitudes, with episodes where 180 K
are observed at 80 km altitude in the NH and up to 100
km in the SH. The thermal impact is clearly observed
during an extended decay phase up to about Ls 250◦ . The
impact of a smaller scale “regional” dust storm which
occurred around Ls 320◦ , near the end of the period
studied in this work, is less clear in our data, perhaps
due to inadequate sampling.

Figure 3: Envelope of all the retrieved temperature profiles
(left-hand panel) compared to the a priori data, from the
LMD-MGCM (right-hand panel). The colors in each scan indicate the Solar Longitude. The thick dashed lines represent
the extreme temperatures at any time and place on Mars, after
[11].

Atmospheric densities
Figure 5 shows the global distribution of all the CO2
densities obtained at an altitude of 70 km, together with
the GCM variability at that altitude. There are clear
variations associated with the latitudinal change at the
tangent point. In both hemispheres the sign of the response is similar: as the orbit approaches lower latitudes
the density increases, following the warmer troposphere
at low latitudes. Averaging over these variations one
can discern a seasonal trend with a maximum around
the summer in the SH, around Ls 270◦ , when the atmospheric density in Mars presents a maximum. However,
in Figure 5 the maximum is observed around Ls 200◦ ,
as a response to the MY34 GDS.
This analysis is complementary and agrees with a
similar study of density variability using NOMAD data

Figure 4: Distribution of all the retrieved temperatures with
latitude and solar time throughout MY34, split in the two
hemispheres (NH in the left hand side panels and SH in the
right hand side panels). The top panels show the latitudes and
the Local Solar Time of the observations. The center panels
show the temperatures, and the lower panels the difference between the NOMAD retrievals and the GCM (a priori).

Figure 5: Global distribution of the retrieved CO2 densities for
MY34, split in the two hemispheres (NH in panels on the left
and SH on the right). The top panels show the latitudes and
the Local Solar Time of the observations and the central panels show the densities in log scale. The bottom panels show
the density variation at 70 km altitude through the mission, including latitudinal and seasonal variations (see text); red dots:
NOMAD densities, blue dots: GCM data.

by [14], using another diffraction order, targeting a different observational period (MY35 and MY36, excluding MY34) and focused on mesospheric altitudes. Although devoted to a non-GDS year, their density absolute
values at 70 km agree well between with our results for
MY34.
Local time variability This type of studies is possible using solar occultation data if one takes care of excluding very high latitudes and search for proper times
during the TGO mapping. Here we grouped the results for latitudes 30◦ N-60◦ N and for solar Longitude
Ls 290◦ -310◦ and examined their changes with the local
time. Figure 6 shows the differences between morning
and evening data, including the LMD GCM profiles for
the same scans. The NOMAD profiles are more wavy
than in the model, with a possibly upward propagating
wave with large amplitudes, around 30 K in the mesosphere, which is much weaker in the GCM. The warm
layer at 80 km in the NOMAD morning data could be
related to the recent finding of a warm layer at precisely
that altitude at nighttime in stellar occultation data from
IUVS/Maven [9].
A similar warm layer was reported also by [1] using
ACS solar occultation data in the ACS/MIR channel.
These data are not exactly collocated with NOMAD
observations, but combined with our results, we claim
that there seems to be a peculiar period of time around

Figure 6: Local Time variations in a location box defined Latitudes 30◦ N-60◦◦ N and Ls 290◦ -310◦ , to minimize latitudinal
and seasonal effects. Blue and read lines are morning and
evening data, respectively.

REFERENCES

perihelion, at the end of the Southern summer where
thermal tides at mid latitudes seem to present very strong
amplitudes, greatly affecting the lower mesosphere.
Acknowledgments
The IAA/CSIC team acknowledges financial support
from the State Agency for Research of the Spanish MCI
through the ‘Center of Excellence Severe Ochoa’ award
(DEV-2017-0709) and funding by grant PGC2018-101836B-100 (MCI/AEI/FEDER, EU). FGG is funded by the
project RTI2018-100920-J-I00 (MCI/AEI/FEDER, EU).
This project acknowledges funding by the Belgian Science Policy Office (BELLS), with the financial and contractual coordination by the ESAU Prod ex Office (PEA
4000103401, 4000121493) as well as by UK Space
Agency through grants ST/V002295/1, ST/V005332/1
and ST/S00145X/1 and Italian Space Agency through
grant 2018-2-HHS.0. US investigators were supported
by the National Aeronautics and Space Administration.
References
[1] D. A. Belyaev et al. Revealing a high water abundance in the upper mesosphere of mars
with acs onboard tgo. Geophysical Research Letters,
48(10):e2021GL093411, 2021. e2021GL093411
2021GL093411.
[2] F. Forget et al. Improved general circulation models of the Martian atmosphere from the surface to
above 80 km. J. Geophys. Res., 104:24,155–24,176,
1999.
[3] S. D. Guzewich et al. Mars science laboratory
observations of the 2018/mars year 34 global dust
storm. Geophysical Research Letters, 46(1):71–79,
2019.
[4] O. Korablev et al. The atmospheric chemistry suite
(acs) of three spectrometers for the exomars 2016
trace gas orbiter. Space Sci. Rev, 214(1):62, Feb
2018.
[5] G. Liuzzi et al. Methane on mars: New insights
into the sensitivity of ch4 with the nomad/exomars
spectrometer through its first in-flight calibration.
Icarus, 321, 09 2018.
[6] M. A. López-Valverde and others. Investigations of the mars upper atmosphere with exomars trace gas orbiter. Space Science Reviews,
214(1):29, Jan 2018. Como citar este articulo:

López-Valverde, M.A., Gerard, JC., GonzálezGalindo, F. et al. Space Sci Rev (2018) 214: 29.
https://doi.org/10.1007/s11214-017-0463-4.
[7] M. A. López-Valverde et al. Martian atmospheric
temperature and density profiles during the 1st
year of nomad/tgo solar occultation measurements.
JGR-Planets, submitted, 2022.
[8] L. Montabone et al. Martian year 34 column
dust climatology from mars climate sounder observations: Reconstructed maps and model simulations. Journal of Geophysical Research: Planets,
125(8):e2019JE006111, 2020.
[9] H. Nakagawa et al. A warm layer in the nightside
mesosphere of mars. Geophysical Research Letters,
47(4):e2019GL085646, 2020. e2019GL085646
2019GL085646.
[10] T. Navarro et al. Detection of detached dust layers in the martian atmosphere from their thermal
signature using assimilation. Geophysical Research
Letters, 41(19):6620–6626, 2014.
[11] M. D. Smith et al. Thermal Structure and Composition, chapter 4, pages 42–75. Cambridge Planetary
Science. Cambridge University Press, 2017.
[12] G. P. Stiller, editor.

The Karlsruhe Optimized
and Precise Radiative Transfer Algorithm (KOPRA),
volume FZKA 6487 of Wissenschaftliche Berichte.

Forschungszentrum Karlsruhe, 2000.
[13] I. R. Thomas et al. Calibration of nomad on esa’s
exomars trace gas orbiter: Part 1 ? the solar occultation channel. Planetary and Space Science, page
105411, 2021.
[14] L. Trompet et al. Carbon dioxide retrievals from
nomad-so on esa?s exomars trace gas orbiter and
temperature profiles retrievals with the hydrostatic
equilibrium equation. i. description of the method.
JGR-submitted, 2022.
[15] A. C. Vandaele et al. Nomad, an integrated suite
of three spectrometers for the exomars trace gas
mission: Technical description, science objectives
and expected performance. Space Science Reviews,
214(5):80, 6 2018.
[16] T. von Clarmann et al. Retrieval of temperature and
tangent altitude pointing from limb emission spectra recorded from space by the Michelson Interferometer for Passive Atmospheric Sounding (MIPAS). J. Geophys. Res., 108(D23), 2003.