A COMPARISON BETWEEN CRISM AND CHEMCAM PASSIVE SKY
WATER VAPOR RETRIEVALS OVER GALE CRATER
A. SJ. Khayat1,2 (alain.khayat@nasa.gov), McConnochie1,3, T., and M. D. Smith1. 1NASA Goddard Space
Flight Center, Greenbelt, MD, USA,2Center for Research and Exploration in Space Science & Technology
(CRESST II), Department of Astronomy, University of Maryland, MD, USA,3Space Science Institute, Boulder,
CO, USA.

Background
Spacecraft observations provide a global view of
the Martian atmosphere by monitoring its behavior
across the planet. In contrast, landers and rovers provide information on the local environment around
their landing sites.
The Mars Reconnaissance Orbiter (MRO) has
been making observations since November 2006. A
key objective of the Compact Reconnaissance Imaging Spectrometer for Mars (CRISM) on MRO
(Murchie et al., 2007) is the “atmospheric monitoring
to characterize the spatial and temporal properties of
the atmosphere for the studies of climate, weather and
photochemistry”. We report atmospheric water vapor
column abundances at a spatial resolution of ~ 1 km
from observations taken over Gale crater, the landing
site of the Mars Science Laboratory (MSL). The 61
targeted mode observations shown in Figure 1 span a
little more than 3 Mars years (MY) from MY 28 at Ls
= 116° (October 04, 2006) to MY 31 at Ls = 101°
(April 24, 2012). The seasonal cycle of water vapor
above Gale is compared against that from reprocessed
MSL Curiosity rover ChemCam passive sky observations using updated gas absorption parameters and
covering the period between MY 31 at Ls = 291°
(March 30, 2013) and MY 35 at Ls = 42° (June 20,
2019.

Figure 1: CRISM targeted observations used in this work
overlaying a topography map by the Mars Orbiter Laser Altimeter. Most of the observations were taken around the
MSL landing site indicated by the red ellipse. The regions
covered by the observations are indicated by the shaded areas in purple, yellow and cyan colors for the Half Resolution Long (HRL), Half Resolution Short (HRS) and Full
Resolution Targeted (FRT) observations, respectively, and
annotated with their numbers.

Methodology
CRISM is affected by the “spectral smile”, an artifact caused by the optical distortions onto the instrument’s detector, leading to a non-uniform variation of
the central wavelength and the point spread function
(PSF) of the detector matrix in the across track direction (Ceamanos & Doute, 2010). This effect is minimal around the central columns, commonly referred
to as the “sweet spot”. Previous water vapor abundance retrievals (e.g., Smith et al., 2009; 2018; Khayat et al., 2019) were conducted by binning the central
100 × 100 pixels, covering a central area 2 km × 2 km
in every CRISM image, and others (Khayat et al.,
2020) used pixels in the along-track direction to track
the changes of water vapor abundance over the polar
troughs. We derived the wavelength offset for every
pixel in Figure 2 and used it as an input in the radiative
transfer code to cover a larger portion of every
CRISM observation.

Figure 2: Area on Mars covered by the CRISM observation HRL0000C95A taken on September 19, 2008 at Ls =
129.4°, and centered at longitude 223.12 °W and latitude
4.51°S. The map of the derived wavelength offset shows
an increase in offset in the cross-track dimension, but no
dependence in the along-track direction.

The retrieval algorithm used here was first described in Smith et al. (2009) and subsequently
adapted and used to provide CRISM atmospheric water vapor retrievals over surface ice in the north polar
regions (Khayat et al., 2019, 2020), and to

characterize the climatology of carbon monoxide by
retrieving the seasonal and spatial variations of the
molecule’s mixing ratio (Smith et al., 2018; 2021).
The water vapor column abundance is retrieved by adjusting the water vapor abundance after correcting for
the retrieved wavelength offset at every pixel (see Fig.
2) until the computed spectrum best matches the observed CRISM spectrum.
The retrieval results from ChemCam observations
presented in McConnochie et al. (2018) provided total
column abundance of water vapor until MY 33 at Ls
= 127°. McConnochie reprocessed the data here using
updated versions of the line list parameters from the
HITRAN 2016 database (Gordon et al., 2017) including parameters for the CO2 band near 783 nm and 789
nm that is used a reference in the H2O retrievals and
extended the coverage to include two additional Mars
years until MY 35 at Ls =42° (June 20, 2019).
Results and conclusion
The seasonal trend of water vapor in Figure 3 as
retrieved by CRISM (black diamonds) follows what
is expected over equatorial regions with the timing of
the aphelion minimum and the maximum around
northern autumn at a time when the sublimated water
vapor transported from the north polar region reaches
equatorial regions (e.g., Smith et al., 2009;
McConnochie et al., 2018). At the beginning of northern spring, the scaled water vapor column abundance
over Gale shows values around 7 pr-µm, followed by
a decline in the water abundance reaching values as
low as 3 pr-µm at the driest time of the year over Gale
around mid to late of northern spring at Ls = 60°. Later
in the season the water vapor column abundance follows a steadily rising trend from 3 pr-µm to peak values ~ 15 pr-µm around the end of northern summer
and early northern autumn at Ls = 180°. The water
abundance column abundance shows a steady decline
during northern autumn and winter, reaching values
between 6 pr-µm and 8 pr-µm in the Ls = 315° - 360°
period.

Figure 3: Water vapor column abundance scaled to a surface
pressure of 6.1 mbar. The black diamonds indicate the median value over each observation of the retrieved CRISM
water vapor column from multiple Mars years between MY
26 at Ls = 116° and MY 31 at Ls = 101° conducted at ~ 15:00
LTST. The error bars indicate the standard deviation over
the water columns across every observation. The squares in

orange indicate the updated ChemCam water vapor column
retrievals for observations taken during multiple Mars years
between MY 31 at Ls = 291° and MY 35 at Ls = 42° from
7:15 to 14:24 LTST. The filled red circles represent the seasonal THEMIS dust opacity for each corresponding to the
CRISM observations.

As a result of implementing updated spectral parameters, there is modestly more water vapor in the
new ChemCam retrievals compared to the results
shown in McConnochie et al. (2018). The results in
Figure 3 (orange squares) show some variations during northern spring during the Ls = 15 -70° period,
with retrieved water vapor abundances between 2.5
pr-µm and 8 pr-µm at Ls = 30°. Other variations in the
retrievals are observed between Ls =150° and 270°,
with a maximum range of ~10 pr-µm (between ~7 prµm and ~17 pr-µm) around Ls =180°.
The annual variation of water vapor from both updated retrievals over multiple Mars years shows a
much closer match than before to McConnochie et al.
(2018). A moderate difference occurring in some
cases could likely be attributed to the interannual variations in water vapor and to the difference in the retrieval approaches between CRISM and ChemCam
and in using different spectral bands of water.
References
Ceamanos, X., and Doute, S., 2010. Spectral Smile Correction of CRISM/MRO Hyperspectral Images. IEEE Transactions on Geosc. and Rem. Sens., 48, no. 11, pp. 3951-3959.
Gordon, I.E., et al., 2017. The HITRAN2016 molecular
spectroscopic database, JQRST 203, 3-69.
Khayat, A., et al., 2019. Understanding the Water Cycle
Above the North Polar Cap on Mars Using MRO CRISM
Retrievals of Water Vapor. Icarus, 321: 722 - 735.
Khayat, A. S.J., et al., 2020. Detections of water vapor increase over the north polar troughs on Mars as observed by
CRISM. GRL, 47.
McConnochie, T. et al., 2018. Retrieval of water vapor column abundance and aerosol properties from ChemCam passive sky spectroscopy. Icarus, 307, 294-326.
Murchie, S., R. et al., 2007. Compact Reconnaissance Imaging Spectrometer for Mars (CRISM) on Mars Reconnaissance Orbiter (MRO), JGR 112(E11).
Smith, M.D., et al., 2009. Compact Reconnaissance Imaging Spectrometer observations of water vapor and carbon
monoxide. JGR 114, E00D03.
Smith, M.D., et al., 2018. The climatology of carbon monoxide and water vapor on Mars as observed by CRISM and
modeled by the GEM-Mars general circulation model. Icarus 301,117–131.
Smith, M.D. et al., 2021. The climatology of carbon monoxide on Mars as observed by NOMAD nadir-geometry observations, Icarus, 362.