THE DIURNAL AND SEASONAL VARIATION OF WATER ICE CLOUDS
AS OBSERVED BY EMIRS
S. A. Atwood, Space and Planetary Science Center, and Department of Earth Sciences, Khalifa University, Abu
Dhabi, UAE, Laboratory for Atmospheric and Space Physics, University of Colorado Boulder, Boulder, USA
(Samuel.Atwood@lasp.colorado.edu), M. D. Smith, NASA Goddard Space Flight Center, Greenbelt, MD, USA,
K. Badri, Mohammed Bin Rashid Space Center, Dubai, UAE, C. S. Edwards, N. Smith, Northern Arizona
University, Flagstaff, AZ, USA, P. R. Christensen, S. Anwar, Arizona State University, Tempe, AZ, USA, M. J.
Wolff, Space Science Institute, Boulder, CO, USA, F. Forget, Laboratoire de Météorologie Dynamique, Jussieu, Paris, France, M. R. El-Maarry, Space and Planetary Science Center, and Department of Earth Sciences,
Khalifa University, Abu Dhabi, UAE
Introduction: A considerable record of Martian
water ice cloud observations has now been produced
from the combined measurements of numerous Mars
missions. However, much of this record is derived
from sun-synchronous spacecraft that observe at
roughly the same local time across their domain
(typically early afternoon) or surface instruments
with no spatial coverage beyond the lander location.
Where local time variable observations of water ice
cloud exist, spatial, temporal, or seasonal coverage is
often limited. Such bounds to observational coverage
hinder efforts to develop Mars atmospheric climatologies, validate global circulation models, and resolve
observational anomalies (Clancy et al., 2017).
The Emirates Mars Infrared Spectrometer
(EMIRS)—a thermal-infrared spectrometer onboard
the Emirates Mars Mission (EMM) spacecraft capable of retrieving water ice cloud column optical
depths—observes across all local times and a wide
range of latitudes and longitudes, yielding broad
coverage at relatively short timescales on the order
of two weeks. EMM began science operations in
May 2021 near the beginning of Mars aphelionseason, a period of the Martian year during which
equatorial water ice clouds are prevalent (Smith,
2004; Clancy et al., 2017).
Here, we present initial results of EMIRS water
ice cloud optical depth retrievals focused on diurnal
variability during the aphelion season (nominally Ls
40°–140°). Additionally, we compare EMIRS observations against modeled water ice cloud abundances
from the Laboratoire de Météorologie Dynamique
(LMD) global circulation model (GCM), and discuss
drivers of diurnal variability in both the equatorial
aphelion cloud belt (ACB) and orographic clouds
associated with volcanoes.
EMIRS Data and Methods: The EMIRS thermal-infrared spectrometer observes Mars with a
spectral range of 6.0 to ~100 µm (1666 to ~100 cm-1)
at typical resolutions of 5 or 10 cm-1. The spacecraft
is in a low-inclination, 55-hour period elliptical orbit
that varies from approximately 20,000 to 43,000 km.
Regular raster scans of the Martian disk occur
throughout the orbit, with observation footprint sizes
ranging from ~100–300 km on the surface (Almatroushi et al., 2021; Edwards et al., 2021).

The EMIRS retrieval follows the algorithm described by Conrath et al. (2000) and Smith et al.
(2004), using a constrained linear inversion to sequentially fit the atmospheric temperature profile,
dust and water ice column optical depths, and water
vapor column abundance. Important to this method
is the existence of sufficient thermal contrast between the atmosphere and surface to distinguish between the spectral features of atmospheric constituents. As EMIRS observations span a range of local
times this contrast varies, typically reaching the lowest values near dawn and dusk. Consequently, the
EMIRS retrieval includes fitting of both the 40 µm
(250 cm-1) and 12 µm (825 cm-1) water ice features.
The 40 µm feature occurs in a region of the spectrum
with more observed radiance and therefore higher
signal-to-noise ratio for fitting purposes as compared
to the 12 µm feature, especially at colder temperatures.
EMIRS observational coverage for analyses
shown here began with the start of EMM science
operations at Ls 48° and included all data through Ls
140°. A gap in observations occurred between approximately Ls 100°–120° due to solar conjunction
and a spacecraft safe mode event, during which the
peak in ACB optical depths (typically around Ls
105°) likely occurred. Aphelion-season water ice
optical depth averages were first filtered to remove
spectra with emission angles greater than 70°, or
retrieved surface temperatures below 190 K—
limiting the results shown here largely to daytime
hours. Additional work to retrieve aerosol optical
depths at lower surface temperatures is ongoing.
Results and Discussion: The spatial distribution
of aphelion-season averaged water ice cloud optical
depths is shown for local true solar times (LTST)
between 0700 and 1700 in Figure 1. The ACB was
clearly detectable at all daytime hours, and was generally observed within the band between 10°S and
30°N (black dashed lines) that is often used to bound
the ACB latitudinal extents, particularly during afternoon hours. Longitudinal variability was also evident including higher optical depths over Tharsis and
lower values just east of the prime meridian. These
observations of the ACB were consistent with

previous observations (Smith, 2004, 2009; Clancy et
al., 2017; Wolff et al., 2019; Giuranna et al., 2021),
however, variability in local time indicates somewhat more complex behavior as compared to results
derived primarily from mid-afternoon observations.
Throughout much of the ACB the average diurnal minimum in optical depth occurred near midday,
with higher values often occurring both in the morning and afternoon. Zonal averages produced minima
between 1200–1300 LTST—though spatial differences in the timing of the minimum varied between
0900 and 1500, complicating this simple notion of a
general midday ACB minimum to some extent. Seasonal changes typically involved scaling in magnitude as higher optical depths were observed closer to
the aphelion-season maximum around Ls 105°, but
with less of an effect on the relative diurnal pattern.
The primary exception to this occurred late in aphelion season when early morning clouds tended to
extend further north and south of the nominal ACB
latitude band.
Distinctions in the diurnal behavior of water ice
clouds were detectable in the vicinity of volcanoes
where orographic clouds form due to upslope winds
and lee waves propagating off the high elevation
peaks (Michaels et al., 2006). In the early morning
these clouds were often absent or not detectable
above the more spatially extensive ACB cloud signal. As the morning ACB clouds decreased small
regions of cloud development were observed by

Figure 1 EMIRS aphelion-season averaged water ice
cloud optical depths for six local times. Bins of 2-hr
LTST, 4°x4° latitude-longitude averages were smoothed
by 2-D Gaussian kernel convolution.

Figure 2 Diurnal difference between aphelion-season averaged afternoon and morning optical depth for EMIRS
observations (top) and LMD GCM output (bottom). Blue
(red) indicates morning optical depths were higher (lower)
than afternoon values.

midday over Olympus Mons, Tharsis Montes, and
Elysium Mons. Orographic cloud optical depths continued to increase throughout the afternoon with no
indication of a maxima having been reached by dusk
when diurnal coverage ends in this analysis. Some
indication of similar behavior was also evident over
Alba Patera, though with more limited observational
coverage making the signal less evident.
Further investigation of spatial differences in diurnal behavior was conducted through a simple parameterization of the difference in average optical
depth between morning and afternoon, shown for
regions with sufficient diurnal coverage in Figure 2.
Regions with higher optical depths in the morning as
compared to the afternoon (blue colors) were particularly evident over the Tharsis region, where early
morning clouds extended beyond the nominal
bounds of the ACB. A latitudinal extension of water
ice cloud beyond afternoon ACB observations has
been discussed previously (Heavens et al., 2010;
Pankine et al., 2013; Wolff et al., 2019; Guha et al.,
2021). This was particularly evident after Ls 120°,
implying some seasonal variability in ACB diurnal
behavior beyond simple changes to the magnitude of
the optical depth. Conversely, the higher afternoon
optical depths (red colors) seen for orographic volcano clouds were noteworthy for their high contrast
in diurnal signal when compared to the morning
dominated ACB clouds in surrounding regions.
A similar parameterization of differences in water ice mass between morning and afternoon was
generated for LMD GCM output using roughly the
same set of seasonal time periods as the EMIRS observations, allowing for a qualitative comparison to
EMIRS results. The spatial distribution for this parameterized diurnal signal was in general agreement
between EMIRS and LMD GCM output for both
ACB and orographic volcano clouds. Interestingly,
the alternating wave-like pattern of morning and
afternoon peaks throughout the ACB was also broadly similar. Model simulations have attributed much
of the spatial and diurnal patterns in these ACB
clouds to thermal tides and associated topographic
effects (Hinson and Wilson, 2004; Clancy et al.,
2017; Szantai et al., 2021). Qualitative differences
were present in some regions, particularly north of
Hellas around 20°N latitude where the LMD GCM
was higher in the early morning as compared to
EMIRS observations. Given that morning observations are generally less available in the observational
record of Martian water ice clouds, additional work
to investigate the both the retrieval and model behavior in this region and local time may be of interest.
Finally, among previously reported local time
variable observations of aphelion water ice cloud,
similar trends in optical depths have been noted for
increasing afternoon ACB optical depths (Benson et
al., 2003; Smith, 2019), midday ACB minima (Akabane et al., 2002; Tamppari et al., 2003; Giuranna et
al., 2021), and higher afternoon values for orograph-

ic volcano cloud regions. The additional spatial, diurnal, and seasonal coverage provided by EMIRS
extends this observational record and adds detail to
the current understanding of diurnal trends in Mars
aphelion water ice cloud development.
Acknowledgements: SAA is supported by the grant
(8474000332-KU-CU-LASP Space Sci.)
References:
Akabane, T. et al. (2002) ‘Diurnal variation of Martian
water-ice clouds in Tharsis region of the low latitude
cloud belt: Observations in 1995–1999 apparitions’,
Astronomy & Astrophysics, 384(2), pp. 678–688.
doi:10.1051/0004-6361:20020030.
Almatroushi, H. et al. (2021) ‘Emirates Mars Mission
Characterization of Mars Atmosphere Dynamics and
Processes’, Space Science Reviews, 217(8), p. 89.
doi:10.1007/s11214-021-00851-6.
Benson, J.L. et al. (2003) ‘The seasonal behavior of water
ice clouds in the Tharsis and Valles Marineris regions
of Mars: Mars Orbiter Camera Observations’, Icarus,
165(1), pp. 34–52. doi:10.1016/S00191035(03)00175-1.
Clancy, R.T. et al. (2017) ‘Mars Clouds’, in The Atmosphere and Climate of Mars. Cambridge University
Press.
Conrath, B.J. et al. (2000) ‘Mars Global Surveyor Thermal
Emission Spectrometer (TES) observations: Atmospheric temperatures during aerobraking and science
phasing’, Journal of Geophysical Research: Planets,
105(E4), pp. 9509–9519. doi:10.1029/1999JE001095.
Edwards, C.S. et al. (2021) ‘The Emirates Mars Mission
(EMM) Emirates Mars InfraRed Spectrometer
(EMIRS) Instrument’, Space Science Reviews,
217(7), p. 77. doi:10.1007/s11214-021-00848-1.
Giuranna, M. et al. (2021) ‘The current weather and climate of Mars: 12 years of atmospheric monitoring by
the Planetary Fourier Spectrometer on Mars Express’,
Icarus, 353, p. 113406.
doi:10.1016/j.icarus.2019.113406.
Guha, B.K., Panda, J. and Wu, Z. (2021) ‘Observation of
aphelion cloud belt over Martian tropics, its evolution, and associated dust distribution from MCS data’,
Advances in Space Research, 67(4), pp. 1392–1411.
doi:10.1016/j.asr.2020.11.010.
Heavens, N.G. et al. (2010) ‘Water ice clouds over the
Martian tropics during northern summer’, Geophysical Research Letters, 37(18).
doi:10.1029/2010GL044610.
Hinson, D.P. and Wilson, R.J. (2004) ‘Temperature inversions, thermal tides, and water ice clouds in the Martian tropics’, Journal of Geophysical Research: Planets, 109(E1). doi:10.1029/2003JE002129.
Michaels, T.I., Colaprete, A. and Rafkin, S.C.R. (2006)
‘Significant vertical water transport by mountaininduced circulations on Mars’, Geophysical Research
Letters, 33(16). doi:10.1029/2006GL026562.
Pankine, A.A. et al. (2013) ‘Retrievals of martian atmospheric opacities from MGS TES nighttime data’, Icarus, 226(1), pp. 708–722.
doi:10.1016/j.icarus.2013.06.024.

Smith, M.D. (2004) ‘Interannual variability in TES atmospheric observations of Mars during 1999–2003’, Icarus, 167(1), pp. 148–165.
doi:10.1016/j.icarus.2003.09.010.
Smith, M.D. (2009) ‘THEMIS observations of Mars aerosol optical depth from 2002–2008’, Icarus, 202(2),
pp. 444–452. doi:10.1016/j.icarus.2009.03.027.
Smith, M.D. (2019) ‘Local time variation of water ice
clouds on Mars as observed by THEMIS’, Icarus,
333, pp. 273–282. doi:10.1016/j.icarus.2019.06.009.
Szantai, A. et al. (2021) ‘Martian cloud climatology and
life cycle extracted from Mars Express OMEGA
spectral images’, Icarus, 353, p. 114101.
doi:10.1016/j.icarus.2020.114101.
Tamppari, L.K., Zurek, R.W. and Paige, D.A. (2003) ‘Viking-era diurnal water-ice clouds’, Journal of Geophysical Research: Planets, 108(E7).
doi:10.1029/2002JE001911.
Wolff, M.J. et al. (2019) ‘Mapping water ice clouds on
Mars with MRO/MARCI’, Icarus, 332, pp. 24–49.
doi:10.1016/j.icarus.2019.05.041.