The Martian oxygen red line airglow: observations with NOMAD/UVIS and
modeling
J.-C. Gérard, LPAP, STAR Institute, Université de Liège, Belgium (jc.gerard@uliege.be), L. Soret, LPAP,
STAR Institute, Université de Liège, Belgium, L. Gkouvelis, L. Gkouvelis, NASA Ames Research Center, CA,
United States, S. Aoki, University of Tokyo, Japan, and the NOMAD/UVIS team
Detection of visible emissions. The Ultraviolet and
Visible channel (UVIS) of the NOMAD instrument
(Vandaele et al., 2018; Patel et al., 2017) on board
the EXOMARS Trace Gas Orbiter (TGO) has been
observing the dayside limb of the planet since April
2019. For this purpose, the TGO spacecraft in moved
in such a way that the nadir channel of the instrument is oriented to the sunlit limb away from the
Sun’s direction. The instrument is occasionally operated to collect spectra in the inertial limb mode
where the instrument points at the dayside limb in a
fixed orientation (López-Valverde et al., 2018). By
virtue of the spacecraft motion along its orbit, the
line of sight scans through the atmosphere down to a
predetermined altitude of the tangent point and detects ultraviolet and visible dayglow emissions. In
this inertial tracking mode, only two altitude scans
through the atmosphere (one ingress and one egress)
are collected during one TGO orbit. In contrast, in
the limb tracking mode, the UVIS boresight is constrained to remain in the vicinity of a preset tangent
altitude while the tangent point of the line of sight
moves in latitude under the effect of the orbital motion. Actually, the tangent altitude usually stabilizes
within ∼20 km during the course of an observation
sequence. Different fixed altitudes have been selected to optimize the observing time near the peak
altitude. Only spectra collected when the solar zenith
angle at the tangent point was less than 70° are used
in this study (Soret et al., 2022).
The 630-nm emission is too weak to determine
limb profiles based on spectra collected during a
single TGO orbit. Therefore, we have averaged the
observed limb distribution of the global dataset and
determined the intensity inside 20 km wide altitude
bins.

The red line feature is observed in all altitude
bins, confirming the reality of its detection. It shows
a peak much broader than other emissions such as
the CO2+ UV doublet or the [OI] green line. A peak
intensity of ~4.8 kR is observed in the 120-140 km
bin of tangent altitude. The large values of the uncertainty compared to the mean intensity are a consequence of both the low observed brightness and
the true variability largely due to the combination of
observations at different latitudes, solar longitudes
and solar zenith angles.
The red line doublet. The O (1D → 3P) transition
giving rise to the red emission is a doublet with a
second, weaker, component at 636.4 nm. This second component is a factor of 3 to 3.1 weaker than
the one at 630 nm (Fischer and Tachiev, 2004;
Sharpee and Slanger, 2006). Although it was not
initially detected, accumulation of spectra now
makes it possible to observe the second doublet
component (Figure 2). We observe a I(630
nm)/I(636.4 nm) intensity ratio in full agreement
with the expected ratio, definitely confirming the
identification.

Figure 2: average UVIS-IUVS spectrum showing
the red line doublet. The horizontal dashed lines
show the 0 signal level and ±1σ noise level (Soret et
al., 2022).
Model.

Figure 1: UVIS mean limb spectra averaged in 20km altitude bins. The insert is a zoom on the 620–
640 nm region. The dotted line indicates the 630 nm
wavelength position (Gérard et al., 2021).

We use the Photochemical Airglow Mars (PAM)
model (Gkouvelis et al., 2018; Gérard et al., 2021;
Soret et al., 2022) to calculate the production, loss
and brightness of the red line dayglow. Figure 3
shows the altitude distribution of the O(1D) main
sources. The production processes are similar to
O(1S). However, the O(1D) density peaks significantly higher than O(1S) because its radiative lifetime is
more than two orders of magnitude longer and collisional quenching plays a major role. Unlike O(1S)

(Gérard et al., 2020) the simulated limb intensity
shows a single peak near 140 km, in agreement with
the observations.

Figure 3: volume production rate of O(1D) atoms in
the dayglow. The individual contributions are indicated in different colors and the total production
rate corresponds to the black dashed line.
We also present the observed seasonal variations of
the 630-nm airglow and compare them with other
dayglow emissions, including the OI 557.7 nm emission.
References
Fischer, C. F., & Tachiev, G. (2004). Atomic Data
and Nuclear Data Tables, 87(1), 1-184.
Gérard, J. C., et al. (2020). Nature Astronomy, 4(11),
1049-1052.
Gérard, J.C. et al. (2021). Geophysical Research
Letters, 48(8).
Gkouvelis, L. et al., (2018). Journal of Geophysical
Research: Planets, 123(12), 3119-3132.
López-Valverde, et al. (2018). Space Science Reviews, 214(1), 29.
Patel, M. R. (2017). Applied optics, 56(10), 27712782.
Sharpee, B. D., & Slanger, T. G. (2006). The Journal
of Physical Chemistry A, 110(21), 6707-6710.
Soret et al. (2022). Submitted for publication.
Vandaele, A. C. (2018). Planet. Space Sci 119, 233249.