RETRIEVAL OF CO COLUMN ABUNDANCE IN THE MARTIAN
THERMOSPHERE FROM FUV DISK OBSERVATIONS BY EMM EMUS
J. S. Evans, J. Correira, Computational Physics, Inc (CPI), Springfield, VA, USA (evans@cpi.com), J. Deighan,
S. Jain, Laboratory for Atmospheric and Space Physics (LASP) at University of Colorado, Boulder, CO, USA,
H. Almatroushi, Mohammed Bin Rashid Space Center (MBRSC), Dubai, UAE, H. Almazmi, United Arab
Emirates Space Agency, Abu Dhabi, UAE, M. Chaffin, Laboratory for Atmospheric and Space Physics (LASP)
at University of Colorado, Boulder, CO, USA, S. England, Virginia Polytechnic Institute and State University,
Aerospace and Ocean Engineering, Blacksburg, VA, USA, M. Fillingim, Space Sciences Laboratory (SSL),
University of California, Berkeley, CA, USA, F. Forget, Laboratoire de Météorologie Dynamique, Institut Pierre
Simon Laplace Sorbonne Université, Paris, France, G. Holsclaw, Laboratory for Atmospheric and Space Physics
(LASP) at University of Colorado, Boulder, CO, USA, R. Lillis, Space Sciences Laboratory (SSL), University of
California, Berkeley, CA, USA, F. Lootah, Mohammed Bin Rashid Space Center (MBRSC), Dubai, UAE.

Introduction:
Carbon monoxide (CO) plays a major role in the
chemical cycles of carbon dioxide (CO2), hydrogen,
and oxygen, and is a tracer of the thermal profile and
winds in the Martian middle atmosphere. With a mean
lifetime of about 6 Mars years (Krasnopolsky, 2007),
CO is produced primarily through CO2 photolysis in
the middle atmosphere and is effectively lost in the
lower atmosphere through catalysis by the OH radical
produced by water vapor (H2O) photolysis. Given
such a long lifetime (relative to one Mars sol), no
variation of CO with local time is expected. However,
as an incondensable, CO is expected to vary both
locally and temporally with Mars season. The spatial
distribution of CO is therefore an important quantity
that can be used to constrain models of photochemical
and dynamical processes in the Martian atmosphere.
From its vantage point in orbit, the Emirates
Ultraviolet Spectrometer (EMUS) onboard the
Emirates Mars Mission (EMM) Hope probe (Amiri et
al., 2021) images Mars at extreme- and far-ultraviolet
wavelengths extending from approximately 100 to
G.M. Holsclaw
al.
170 nm. The EMUS ΣCO/CO2 algorithm
(whereet Σ

indicates a slant path integral) provides a measure of
relative composition variability within the Martian
thermosphere by deriving the column abundance of
CO above a fixed reference column density of CO2
from EMUS images. The ΣCO/CO2 algorithm traces
its origins to the ΣO/N2 algorithm used at Earth for
studying relative variability in atomic oxygen
abundance from remote sensing observations
(Strickland et al., 1995; Evans et al., 1995; Correira
et al., 2021). In a similar manner, relative composition
can be inferred for the Martian atmosphere using CO
and CO2.
Observations:
We use newly released data taken with EMUS
aboard the EMM Hope probe to infer relative
composition variability within the Martian
thermosphere by deriving the column abundance of
CO above a fixed reference column density of CO2.
The orbit of the EMM Hope probe has an apoapsis at
42,650 km and periapsis at 19,970 km with a 54.5
hour period. This high altitude orbit affords a synoptic

tion
ms on the
of the
square
ight
rs and the
of
rams on
servation
of the
ea scanned
ument
ngles)
regions of
details)

(c)

HI

OI
OI
CO 4PG
CI
CI

(d)

Figure 1. (a) Left: Position of EMM in its orbit (square symbols) and lines of sight (arrows) relative to Mars and
the Sun for U-OS1 observations. Right: Observation mode from the perspective of EMM, with the area scanned
by the 10.75° tall instrument slit (grey rectangles) overlaid on Mars. (b) Same as (a) but for U-OS2 observations.
(c) EMUS spectrum measured on April 24, 2021 shown in black with multiple linear regression (MLR) fit shown
in red. (d) The difference (counts/bin) between the measured spectrum and MLR total fit.

1

view of Mars with full local time coverage every 9-10
days (Holsclaw et al., 2021). High and low resolution
slits have angular widths of 0.18° and 0.25° and
spectral widths of 1.3 nm and 1.8 nm, respectively
(Holsclaw et al., 2021). EMUS remotely senses the
thermosphere at 100 – 200 km altitude, observing farultraviolet (FUV) emissions (see Figure 1) from
hydrogen (H I 121.6 nm), oxygen (O I 130.4 and
135.6 nm), and carbon monoxide (CO 4PG 140 – 170
nm). The first data release, covering February 10,
2021 to May 23, 2021, was delivered in early October
2021. For the present analysis, we utilize EMUS UOS1 and U-OS2 disk observations covering February
20, 2021 to March 18, 2022.

latter uncertainty does not apply to the input
radiances). A rigorous propagation of errors is used,
beginning with the intensity uncertainties present in
Level 2B data.

Figure 2. Solar zenith angles (left) and solar Lyman α
pumped CO 4PG intensity (right) for the second swath
(out of three) for U-OS2 observations on April 24,
2021 (top row) and June 18, 2021 (bottom row).

Retrieval Algorithm:
The ΣCO/CO2 algorithm uses a lookup table
approach by first generating a series of model
atmospheres that span the expected range of
physically realistic atmospheres. When used with
EMUS observations, the lookup table provides a
unique mapping from intensity to CO/CO2 column
density ratio for a given solar zenith angle (SZA) and
emission angle (EMA). Intensities from the optically
allowed CO Fourth Positive Group (4PG) band
system (A1Π → X1Σ+) are used as a signature of CO
variability.
Results:
EMUS Level 3 data files contain ΣCO/CO2
derived from EMUS observations. The algorithm is
primarily intended for U-OS1 observations, though it
may be applied to U-OS2 observations as well, or any
EMUS observation that views the disk of Mars. We
note that the spatial resolution for other EMUS
observation types is typically too coarse to derive
useful data products. U-OS1 is comprised of two
scans, each scan imaging a little more than half the
Martian disk visible from EMUS’ vantage point. UOS2 observations consist of three scans, with one scan
covering approximately 2/3 of the Martian disk, while
the first and last scans cover the edges of the disk (see
Figure 1). Before ingestion by the ΣCO/CO2
algorithm, Level 2 data may be spatially binned to
increase signal to noise.
Figures 2 and 3 present EMUS data. Figure 2
shows the SZA and the intensity of CO 4PG excited
by solar Lyman α photons on the left and right,
respectively. The upper row shows values for the
central swath of an U-OS2 observation from April 24,
2021, while the bottom row shows the central swath
from an U-OS2 observation on June 18, 2021. The left
column of Figure 3 shows ΣCO/CO2 derived from the
algorithm, while the right hand column shows the
percent difference of each pixel from the mean swath
value. Uncertainty values are reported for the input
intensities and derived ΣCO/CO2 and are separated
into random, systematic, and model uncertainties (the

Figure 3. Retrieved ΣCO/CO2 (left) and percent
difference from swath averaged ΣCO/CO2 (right) for
the same observations shown in Figure 2.
References:
Amiri, H.E. S., et al., The Emirates Mars Mission.
Space Sci. Rev, 218, 4 (2021).
Correira, J., Evans, J. S., et al., Thermospheric
Composition and Solar EUV Flux From the GlobalScale Observations of the Limb and Disk (GOLD)
Mission. J. Geophys. Res., 126 (12), e29517 (2021).
Evans, J. S., et al., Satellite remote sensing of
thermospheric O/N2 and solar EUV. 2: Data analysis.
J. Geophys. Res., 100, 12227 (1995).
Holsclaw, G. M., et al., The Emirates Mars
Spectrometer (EMUS) for the EMM Mission. Space
Sci. Rev, 217, 79 (2021).
Krasnopolsky, V. A. Long-term spectroscopic
observations of Mars using IRTF/CSHELL: Mapping
of O2 dayglow, CO, and search for CH4. Icarus, 190,
no. 1 (2007).
Strickland, D. J., Evans, J. S., et al., Satellite
remote sensing of thermospheric O/N2 and solar
EUV. 1: Theory. J. Geophys. Res., 100, 12217 (1995).

2