WATER TRANSPORT IN THE MARTIAN NORTHERN WINTER POLAR
ATMOSPHERE
D. J. McCleese, Synoptic Science, Altadena, CA, H. E. Gillespie, Lunar and Planetary Institute, Houston, TX
(hgillespie@lpi.usra.edu), A. Kleinböhl, D. M. Kass, Jet Propulsion Laboratory, California Institute of Technology, Pasadena, CA, S. J. Greybush, Pennsylvania State University, University Park, PA, R. J. Wilson,
Ames Research Center, Mountain View, CA

Introduction:
The Martian polar vortices are challenging to
study in part due to the large gradients in temperature, wind, and aerosol fields at their boundaries as
well as the transient and stationary waves that often
have their largest amplitudes near the polar vortex
boundaries. This work focuses on water transport in
the vicinity of, and particularly at the boundary of,
the northern winter polar vortex in Mars Year (MY)
30, as observed by the Mars Climate Sounder (MCS)
[1] and simulated by the Ensemble Mars Atmosphere
Reanalysis System (EMARS) [2]. Understanding of
cross-boundary transport is crucial to understanding
the Martian water cycle, particularly the influence of
the multiple kilometers thick permanent north polar
water ice cap. During the northern winter, water
from most of Mars is transported by the Hadley
circulation to the winter pole, where it is deposited
onto the north polar ice cap. Several prior studies
have investigated transport across the polar vortex
boundary, including for varying obliquity over Martian climate cycles [3] and using the “age-of-air”
diagnostic [4]. Here, we investigate transport across
a dynamic polar vortex boundary as a function of
altitude, longitude, and Ls.

Datasets:
MCS is an infrared radiometer aboard the Mars
Reconnaissance Orbiter. MCS observes the atmosphere of Mars at approximately 3 AM and 3 PM
local time and has observed the atmosphere of Mars
since MY 28. Vertical profiles of temperature, dust
opacity, and water ice opacity are retrieved from
MCS radiances [5], and extend from the surface to
about 80 km above the surface, with a vertical resolution of about 5 km.
EMARS is a reanalysis that combines the Geophysical Fluid Dynamics Laboratory Mars Global
Climate Model (MGCM) with MCS observations
using data assimilation, specifically the local ensemble transform Kalman filter (LETKF) [6]. Data assimilation enables EMARS to have full coverage in
longitude, latitude, altitude, and time of any variables of interest in the Martian atmosphere, as models
do, while retaining information about Martian
weather from observations. EMARS assimilates
vertical profiles of temperature from MCS, and the
assimilation updates temperature, horizontal wind,

and surface pressure. Dust is prescribed in the lowest
model levels to conform to column dust opacities in
[7], representing lifting and sedimentation of dust in
the boundary layer. The grid spacing of EMARS is
5° latitude x 6° longitude, with 28 vertical levels
extending from the surface to about 100 km above
the surface on hybrid σ-p surfaces. The EMARS
control run, hereafter called the “free run,” is a run
of EMARS without data assimilation. A comparison
of EMARS with its free run allows us to determine
the effects of assimilation on the modeled Martian
atmosphere, and comparison with MCS when possible enables us to assess how close EMARS is to the
true state of the Martian atmosphere.

Figure 1. Zonal averages of temperature retrieved
from MCS measurements (top) compared with
EMARS (middle) and the free run (bottom). The left
three panels are Ls 240ᵒ-245ᵒ; right three panels are
Ls 270ᵒ-275ᵒ.

Behavior of the Polar Vortex:
We begin our analysis of the northern winter polar vortex in MY 30 by examining zonal mean temperature, dust, and water ice over the pole in MCS,
EMARS, and the free run (Figure 1). In Figure 1, the
left side of each panel shows values at 3 AM local
time, while the right side shows values at 3 PM local
time. The vortex is only shown here for 5 Ls intervals following Ls 240 and Ls 270, but qualitatively,
the fields shown in Figure 1 persist from Ls 225 to
Ls 315 with little change. In all cases, the maximum
temperatures in the midlatitudes increase in altitude
(decrease in pressure) from about 300 Pa at 30 N to
about 30 Pa at 60 N. In the polar region, the maxi-

mum temperature increases in altitude more rapidly
when approaching the pole, with the maximum in
temperature at about 0.3 Pa over the pole. EMARS
improves the slope of the maximum thermal gradient
between the polar vortex and the midlatitudes,
whereas the free run vortex is tall and narrow, comparatively. The horizontal temperature gradient is
key to understanding the polar vortex boundary on
its own as well as through its connection to thermal
wind balance and potential vorticity. Temperatures
are generally improved in EMARS compared to the
free run, with the notable exception of the interior of
the polar vortex at Ls 270.

substantially. None of the definitions of the polar
vortex work well near the surface, so we assume that
the lowest 1 km of the vortex has a vertical boundary. Specifically, a grid cell of EMARS or the free run
is in the vortex if these 5 conditions apply:
1) Its temperature is below 170 K
2) No cell closer to the north pole at the same
longitude, height, and time is above 170 K in temperature
3) A cell further from the north pole is above 170
K in temperature
4) It is not on the topmost two model levels
5) It is below 1 km altitude and the cell directly
above it at 1 km is in the vortex.

Figure 3. Zonal mean streamfunction, and other
quantities, in the free run (left) and reanalysis (right)
averaged over 5 sols after winter solstice.
Streamfunction (108 kg/s) is in black, water vapor
mass mixing ratio (10-6 kg/kg) is in red, water ice
mass mixing ratio (also ppm) is in white, and temperature is in color. The 170 K contour is marked in
bold gray.

Figure 2. The upper panel is zonal mean temperature
from EMARS with the vortex boundary bolded red;
the lower panel is the scaled potential vorticity following [8]. The units for the lower panel are scaled
PVU (10-6m2s-1kg-1K).
In order to assess transport in and out of the polar
vortex, we must adopt a definition of the polar vortex boundary. The polar vortex boundary has been
defined in a variety of ways, including the maximum
meridional potential vorticity gradient and the maximum meridional temperature gradient. Here, we
choose a simple, observable definition that roughly
coincides with previous definitions of the polar vortex for Mars, as shown in Figure 2: the 170 K isotherm, with a few modifications. Note that scaled
potential vorticity, defined following [8], reduces the
vertical gradient of potential vorticity without affecting the horizontal gradient of potential vorticity

Water Transport in the Northern Winter Polar Vortex:
Water transport near the northern polar vortex
differs substantially between EMARS and its free
run. In particular, substantially more water enters the
polar vortex in EMARS than in the free run. Figure
3, in particular the zonal mean streamfunction, is
indicative of this increase in transport. In the free
run, lines of constant streamfunction do not cross the
polar vortex boundary above 100 Pa, and substantial
cross-boundary transport occurs only below 300 Pa.
Lines of constant streamfunction are approximately
parallel to the vortex edge at high altitude, suggesting little to no transport of any quantity into the polar
vortex at high altitude in the free run. However, the
EMARS streamfunction shows substantial crossboundary transport up to at least 10 Pa, with lines of
constant streamfunction bending into the polar vortex at the vortex boundary. Compounding this difference in the winds, as inferred by the streamfunction,
is the increased availability of water vapor and water
ice at the vortex edge in EMARS above 500 Pa. A
temperature increase of about 10 K near the surface

within and just outside the polar vortex in EMARS
compared to the free run are consistent with adiabatic warming produced by an enhanced, crossboundary Hadley circulation. It seems that the increased cross-boundary transport causes vertically
extended water ice clouds to appear within the
EMARS vortex, which has been a challenging feature of MCS observations to reproduce with Mars
models [9]. This cross-boundary transport in
EMARS persists not only through nearly the entirety
of the northern winter, but also does not vary substantially among Martian years lacking a global dust
storm.

Figure 4. EMARS (left) and free run (right) simulated cross-sections of mean water transport at four
model levels L7 (10 Pa, 45 km), L10 (36 Pa, 30 km),
L17 (300 Pa, 7 km) and L23 (550 Pa, 1 km) above
the surface for Ls 270ᵒ to 275ᵒ in MY 30. The units
of meridional transport of water are (m/s)*(kg water/kg air)*103.
Figure 4 shows latitude-longitude cross-sections
of mean water (vapor and ice) transport at four model levels in the atmosphere. This figure enables us to
see variation in transport with longitude not visible
in the zonal mean. Notice that transport, as well as
the vortex boundary, are dominated by a wavenumber 1 stationary wave in the upper atmosphere

that transitions to wavenumber 2 in the lower atmosphere. Near-surface transport may result primarily
from katabatic winds from Alba Patera (41°N,
249°E). Also observe that in the free run, water
transport tends to stop at the polar vortex boundary,
while in EMARS, transport continues across the
boundary unimpeded.
Figure 5 explores the vertical dimension of the
total water transport as a time mean across the dynamic vortex boundary between Ls 270 and Ls 300.
In the free run, water is primarily transported into the
vortex at altitudes below the 200 Pa surface.
Transport into the vortex is dominated by vertical
transport, and horizontal transport acts partially in
opposition to the vertical transport, removing some
of the water added to the vortex. The variability of
vertical transport in the free run may be due to the
greater surface area of the vortex at lower altitudes
than 200 Pa compared to high altitudes. In EMARS,
the maximum water transport occurs between 50 and
100 Pa, and comparatively less transport into the
vortex occurs between 200 and 450 Pa, where the
free run transport into the vortex is greatest. The
vertical transport into the vortex is approximately
constant with height, suggesting a vortex edge that is
approximately uniformly sloping with height, as seen
in Figure 4. Variations in transport with height in
EMARS are primarily horizontal. In total, we find
that between Ls 240 and 300, in EMARS, transport
into the vortex is about 6 Tg/sol, with about 2 Tg/sol
occurring at altitudes below the 200 Pa surface. In
the free run, transport into the vortex is less than 4
Tg/sol, and nearly all of it occurs below 200 Pa.

Figure 5. Water transport into the polar vortex in the
EMARS (left) and the free run (right) as a function
of height, in Tg/sol/Pa, averaged from Ls 270 to Ls
300. Total transport is shown in bold red, the
transport due to the horizontal wind is shown in
green, and the transport due to vertical wind is
shown in blue.

Figure 6. Water transport into the polar vortex in
EMARS during Ls 270ᵒ-300ᵒ. Total transport is

shown in bold red, which is split into the amount of
transport into the vortex parallel to isentropes (green)
and perpendicular to the isentropes (blue).
Figure 6 shows the total transport across the vortex boundary, divided into components parallel and
perpendicular to isentropes. This decomposition
assists us in understanding why water is transported
across the vortex boundary. Transport parallel to
isentropes is caused by adiabatic processes (typically), and transport perpendicular to isentropes is
caused by diabatic processes. In EMARS, most of
the transport is cross-isentropic, which is indicative
of diabatic cooling, namely radiative cooling of air in
the descending branch of the Hadley circulation. A
similar result holds for the free run (not shown).
Conclusions:
Our study of water transport shows that the
northern polar vortex is somewhat porous to water.
According to EMARS, about 6 Tg/sol water enters
the polar vortex, and most of the water enters the
vortex above 200 Pa as part of the global Hadley
circulation. The vortex boundary and transport
across it are highly variable in all spatiotemporal
dimensions in EMARS. The free run captures neither
the upper atmosphere transport into the vortex nor
the variability of the vortex boundary and transport
across it. Measurements of wind and additional
measurements of water, particularly water vapor, in
and around the polar vortex with sufficient spatiotemporal coverage to be included in data assimilation
would be helpful in improving the representation of
water transport in Mars simulations as well as in
assessing water transport in the real Martian atmosphere.
Acknowledgments:
Participation by HEG and SJG in this research
has been funded by NASA MDAP awards
80NSSC17K0690 and 80NSSC20K1054. EMARS
and MGCM control run data are publicly available
through the Penn State Data Commons, and access is
described in [2]. Work at the Jet Propulsion Laboratory, California Institute of Technology, was performed under a contract with the National Aeronautics and Space Administration. The Jet Propulsion Laboratory also supported work through contract 1581456. The MCS data set is freely available
from NASA's Planetary Data System (PDS):
https://atmos.nmsu.edu/data_and_services/atmospher
es_data/MARS/mcs.html. ©2021. All rights reserved. Government sponsorship is acknowledged.

References:
[1] McCleese, D. J., et al. (2007). “Mars Climate
Sounder: An Investigation of Thermal and Water
Vapor Structure, Dust and Condensate Distributions
in the Atmosphere, and Energy Balance of the Polar

Regions,” J. Geophys. Res., 112, E05S06,
doi:10.1029/2006JE002790
[2] Greybush, S. J., et al. (2019). The Ensemble
Mars Atmosphere Reanalysis System (EMARS)
Version 1.0, Geosci Data J., 6, 137-150,
doi:10.1002/gdj3.77
[3] Toigo, A.D., D.W. Waugh, and S.D.
Guzewich (2020), Atmospheric transport into polar
regions on Mars in different orbital epochs. Icarus,
347, https://doi.org/10.1016/j.icarus.2020,113816
[4] Waugh, D.W., A.D. Toigo, and S.D.
Guzewich (2018), Age of Martian air: Time scales
for Martian atmospheric transport. Icarus 317,
https://doi.org/10.1016/j.icarus.2018.087.002
[5] Kleinböhl, A., et al. (2009). Mars Climate
Sounder limb profile retrieval of atmospheric temperature, pressure, dust and water ice opacity. J.
Geophys.
Res.,
114,
E10006,
doi:10.1029/2009JE003358
[6] Hunt, B. R., E. J. Kostelich, and I. Szunyogh
(2007), Efficient data assimilation for spatiotemporal
chaos: A local ensemble transform Kalman filter,
Physica
D,
230,
112–126,
doi:
10.1016/j.physd.2006.11.008
[7] Montabone, L., F. Forget, E. Millour, R.J.
Wilson, S.R. Lewis, B. Cantor, D. Kass, A.
Kleinböhl, M.T. Lemmon, M.D. Smith, M.J Wolff
(2015). Eight-year climatology of dust optical depth
on Mars, Icarus, 251, 65-95.
[8] Waugh, D. W., A. D. Toigo, S. D. Guzewich,
S. J. Greybush, R. J. Wilson, and L. Montabone
(2016), Martian polar vortices: Comparison of reanalyses, J. Geophys. Res. Planets, 121, 1770–1785,
doi:10.1002/2016JE005093
[9] Haberle, R.M, M. Kahre, J.L. Hollingsworth,
F. Montmessin, R.J. Wilson, et al., (2019), Documentation of the NASA/Ames Mars Global Climate
Model: Simulations of the present seasonal water
cycle.
Icarus,
https://doi.org/10.1016/j.icarus.2019.03.026