Water vapor, water-ice clouds, and dust in the North Polar Region
L. K. Tamppari, Jet Propulsion Laboratory/Caltech, Pasadena, CA, USA (email@example.com), M.
D. Smith, Goddard Space Flight Center, Greenbelt, MD, USA, D. S. Bass, A. S. Hale, Jet Propulsion Labora-
tory/Caltech, Pasadena, CA, USA
The behavior of water vapor, water-ice and
dust in the Martian atmosphere is important for un-
derstanding the overall Martian climate system,
which is characterized by three main cycles: water,
including water-ice, dust, and CO2. Understanding
these cycles will lend insight into the behavior of the
atmospheric dynamics, the atmosphere’s ability to
transport dust, water-ice, and vapor to different parts
of the planet, and how that ability changes as a func-
tion of dust and water-ice loading.
Because the north polar region has the larg-
est detectable source of water present on Mars today
(Kieffer 1979; e.g., Boynton et al., 2002), examining
water vapor, water-ice clouds and dust in this area is
of interest. Furthermore, understanding their behav-
ior may provide insight into the differences in be-
havior between the polar caps and in the formation
of the polar layered deposits.
Water-ice opacities in the north polar re-
gion are known to be dominated by polar hood
clouds that track the seasonal polar cap in the spring
and late summer/early fall (e.g., Leovy et al., 1972;
Briggs and Leovy, 1974; James et al., 1994), and to
be much smaller in the north polar region than the
equatorial region during summer. This can be seen
clearly in longitudinally averaged maps such as
those provided by Smith (2004, Figure 5). Dust
opacities, on the other hand, tend to be higher in the
north polar region during spring and summer than in
the equatorial regions (e.g., Smith, 2004). This is
due to increased frequency of cap-edge dust storms
(James and Cantor, 2001).
While observations of water-ice clouds and
dust in the Martian atmosphere have been made,
previous studies in the north polar region of Mars
have either focused on imaging data sets (e.g, Wang
and Ingersoll, 2002; James and Cantor, 2001, Cantor
et al., 2001) without optical depth estimates or on
longitudinally averaged infrared data sets which do
provide optical depth estimates (e.g., Smith, 2004;
Liu et al., 2003). Additionally, older data set (pre-
MGS) either did not have full spectral capability
(e.g. Viking Infrared Thermal Mapper [IRTM]) or
did not have multi-year coverage (Mariner 9). Wa-
ter-ice clouds can be detected in the Viking IRTM
instrument (Tamppari et al., 2000, 2003), but have
not yet been retrieved in the polar regions, although
some preliminary latitude-, longitude-, and season-
resolved studies using IRTM have been performed
(Tamppari and Bass, 2000). Furthermore, dust re-
trievals have been performed from the IRTM data
set (e.g., Martin and Richardson, 1993), but these
have been zonally averaged. The MGS spacecraft
provides, for the first time, a data set of regular,
good resolution, long-term coverage from which to
perform such studies.
Smith (2002) examined water vapor in the
north polar region, but only examined longitudinally
averaged data during the first full MGS year. In his
paper, he found that the vapor increase in northern
spring, as the seasonal polar cap is subliming and
sunlight is hitting the region for the first time that
season, appears to be quite rapid in MAWD data, but
is more gradual in TES data (Smith, 2002). Farmer
et al. (1976) observed a water vapor peak in latitude
bands 70-80° at Ls=108. Later MAWD analysis
with a more full data set showed a maximum of 100
pr um between 85-90 N at Ls=120 (Jakosky and
Farmer, 1982). Sprague et al. (2001) obtained 42
measurements higher than 65 N, spanning 1996-
1999. They observed a peak of 76 pr um at about
Ls=113, a little early and a little lower latitude than
MAWD, though roughly consistent. Smith (2002)
shows a maximum of 100 pr m at Ls =115 and 80°
N. Sprague et al. (2001) indicate that their ground
based measurements show a seasonal maximum
much lower than the TES measurements, about 76 pr
m at Ls=113 and 75° N (1998-1999 observations).
However, their measurements from the previous
season, 1996-1997, only show an abundance of ~50
pr m at Ls=114 and 79° N (Sprague et al., 2001).
Water-ice, water vapor, and dust observa-
tions are important for the Phoenix Mars Scout mis-
sion, scheduled to launch in August 2007 and plan-
ning to land in the north polar region between 65-
72N. The Phoenix mission is planning to carry two
experiments that will observe these quantities in the
atmosphere: a solid state imager, developed by the
University of Arizona, and an upward-looking lidar,
contributed by the Canadian Space Agency. For
planning purposes, the amount and behavior of the
vapor, ice, and dust is of interest. Furthermore, the
Phoenix engineering team is also interested in water-
ice and dust opacities. Both water-ice and dust opti-
cal depth will affect the amount of sunlight impin-
gent upon the solar panels, thus modifying the
amount of solar energy present to power the space-
craft and instruments. Dust in the atmosphere will
affect the surface and near-surface atmosphere tem-
peratures, which in turn could affect the amount of
heater power needed to keep the spacecraft warm (it
SHORT TITLE HERE: A. B. Author and C. D. Author
is passively cooled). Finally, the spacecraft entry,
descent and landing will be affected by the amount
of dust present due to its affect on the atmospheric
temperature profile and therefore the density profile
that the spacecraft experiences.
The water vapor and water-ice and dust op-
tical depth mapping done in here utilizes the data
from the MGS TES instrument. It is an infrared
interferometer/spectrometer operating in the spectral
m (Christensen, 1992). In particular,
the water-ice clouds are retrieved using the ~12-m
(825 cm-1) water-ice absorption feature and the dust
is retrieved using the ~9-m (1075 cm-1) (Pearl et
al., 2001; Smith, 2004). Water vapor is retrieved
using 5 bands spanning 28-42 µm.
The MGS spacecraft is in a sun-
synchronous, nearly polar orbit (Christensen, et. al.,
2001). The spacecraft orbits Mars 12 times every 1-
sol period covering the globe with equally spaced
strips once a day. The data are taken around the local
time of 1400 hours and 0200 hours. Here, we use
the daytime (~1400 hour) data. At this time of day,
water-ice clouds formation is likely to be near the
minimum since the diurnal temperatures will be near
the maximum (e.g., Pathak et al., 2004). As such,
water-ice optical depths noted here are likely a lower
limit on the amount present over the diurnal cycle.
Mapping of atmospheric quantities:
The atmospheric quantities mapped here,
using MGS TES observations, span nearly three
Mars spring/summer seasons, from Ls≈104 in Mars
year 24 (1999-2000) to Ls
180 in Mars year 26
(2002-2003; Mars years definitions per Clancy et al.,
2000). Ls is defined as the areocentric longitude of
the sun, with Ls=0 starting at northern spring equi-
nox and stepping through the seasons to Ls=359, just
before the subsequent northern spring equinox. The
data examined cover 60-90°N latitude during the
Mars northern spring and summer times (Ls=0-180).
Because of the low surface temperature in these
northern latitudes, which decreases the signal to
noise, the data examined here are retrieved only over
a surface of T>220K.
To produce seasonal maps, we average the
optical depths and the vapor in boxes of 2º latitude
by 4º longitude by 5º in Ls (Figures 1 and 2 for wa-
ter-ice and dust, respectively and Figure 3 for water
vapor). We chose this combination of parameters to
maximize the areal coverage while minimizing the
time step from map to map, and to allow averaging
of many retrievals together, minimizing the uncer-
tainties in each bin. The mathematical mean is com-
puted for each bin using the total number of points
that fell into that lat/lon/Ls bin. The typical number
of points in any one bin is 10-20, and so we assume
that the uncertainties in each bin are 0.025 or 5% of
the optical depth. The color scale for the maps
shows absorption-only optical depths for τ12μm (wa-
ter-ice) on a scale of 0-0.2 and τ9μm (dust) on a scale
of 0-0.4. For water vapor, the color scal ranges from
0-100 pr microns. The dynamic range of each of the
scale bars was chosen to “stretch” the signal and
bring out variations within the maps. In actuality,
the largest value in a given map is often larger than
the maximum shown and in those cases, is colored to
the maximum color on the scale bar. The black area
in the maps contains no data. The black area sur-
rounding the north pole is due to surface temperature
cutoff of Ts >220K. This area decreases and subse-
quently increases, following the retreat and growth
of the seasonal polar cap.
Maps for all years and seasons will be
shown and discussed at the conference.
Figure 1. TES infrared (12-micron) water-
ice optical depth map for MY26, Ls=95-110. The
color scale bar ranges from 0-0.2.
Figure 2. TES infrared (9-micron) dust op-
tical depth map for MY26, Ls=95-110. The color
scale bar ranges from 0-0.4.
Figure 3. TES water vapor map for MY26,
Ls=95-110. The color scale bar ranges from 0=100
The water-ice cloud and dust optical depths
and water vapor have been mapped in the north po-
lar region during northern spring and summer for
three Mars Years, beginning Ls=105 in MY24 (Mars
Year definitions per Clancy et al., 2000), using the
Mars Global Surveyor Thermal Emission Spec-
trometer (Tamppari et al., 2005). The observations
discussed here are consistent with past work on
clouds and dust using other other techniques (e.g.,
Wang and Ingersoll; 2002, Smith, 2004; Cantor et
Specific conclusions are:
1) Longitudinal variability in the optical
thickness of the NPH exists, both during springtime
recession and late-summer onset. This variability
2) Each year, the breakdown of the sea-
sonal NPH occurs about Ls=70 and the NPH re-
forms near about Ls=160-165. The end of the NPH
did not vary interannually within the data set exam-
ined here, but the onset varied by about 5 degrees.
3) Late spring NPH recession water-ice
clouds and m s
4) Water-ice cloud background opacity
ckground water-ice opacities
6) The observations show many examples
st opacities than other loca-
r 2 systems in the late spring in MY 26,
l constraints for
nsen, P. R., et al., JGR 97, 1992.
002. , 1982
id-summer NPH onset water-ice cloud
tend to be patchy in nature. In other words, they are
confined near the northernmost extent of the data,
but vary significantly in opacity as a function of lon
crease to summertime low levels (typically
<0.1) starting at Ls=80 in each year. The opacities
begin to rise again at different times in different
years: Ls=140 for MY 24, Ls=135 for MY 25, a
Ls=110 for MY 26.
5) The ba
o reach minimum values in the longitudinal
region 0-90 W between Ls=105 and Ls= 125 in all
years, though the onset and retreat of this feature
was present earlier in MY 25 and continued later in
both MY24 and 25. This observation is at the limit
of the uncertainties in the data and cannot be con-
-ice clouds that are associated with elevated
dust opacities as well as examples where water-ice
clouds do not correlate with elevated dust opacities.
Both types are seen at a variety of seasons and loca-
tions. 7) There is significant interannual vari-
ability in dust magnitude, but both MY 25 and 26
show peak dust (high magnitude over wide area)
8) The longitudinal quadrant 0-90 W ex
hibits much higher du
many seasons in all years examined, but the
details vary interannually. Both MY25 and 26
showed this region having the most dust between
9) There is evidence for stationary
rized by elevated dust and water-ice in
opposite quadrants and reduced dust and water-ice in
10) Water vapor varies significantly inter-
annually 11) Water vapor varies spatially within a
season and different years behave differently.
12) Water vapor increases in some loca-
tions above 50 pr microns near Ls=75 and decreases
below 50 pr microns again near Ls=130-125. I
times, there are many locations in which t
column abundance is above 50 pr microns, even
approaching 200 pr microns. This increase/decre
pattern is repeatable year to year.
Because these observations show the spa
and seasonal changes in atmospheric quantities over
3 Mars years, they will provide usefu
f dynamics and water-ice cloud formation.
In addition, the Phoenix Mars Scout Lander, sched-
uled for launch in August 2007, will land between
65-72 N in late May 2008. The mission will be 90
sols long, occurring between Ls=~76-125. Phoenix
will have a stereo camera, with solar filters used for
determining aerosol optical depth and water vapor
abundance and an upward-looking, dual-wavelength,
dual-detector lidar that will be able measure back-
scatter off atmospheric dust and water-ice. Under-
standing the likely behavior of water vapor, water-
ice and dust in the atmosphere will enable better
preparation of Phoenix observational sequences.
References in Alphabetical Order:
Briggs, G. A., and C. B. Leovy, Bull. AMS, 55,
nton, W.V., et al., Science 297, 2002.
Cantor, B., et al., JGR 106(E10), 2001.
Clancy, R. T., et al. JGR 105, 2000.
Christensen, P. R. et al., JGR 106, 2001.
Farmer, C. B., et al, Science 194 (4271),
Feldman, W.C., et al., Science 297, 2
Jakosky, B. M. and C. B. Farmer, JGR 87
SHORT TITLE HERE: A. B. Author and C. D. Author
James PB, and Cantor BA, Icarus 154(1), 2001.
dson, JGR 98(E6),
M.D., Icarus 167, 2004. 1)
The 2nd Mars Po-
et al., submitted to Plan. and Sp.
James, P. B. et al., Icarus 109, 1990.
Kieffer, H. H., JGR 84, 1979.
Leovy et al., Icarus 17, 1972.
Liu JJ, et al., JGR 108 (E8), 2003.
Martin, T. Z., and M. I. Richar
Pathak et al., submitted to Icarus, 2
Pearl et al., JGR 106(E6), 2001
Sprague, A. L. et al., Icarus 154(1), (200
Tamppari, L. K., and D. S. Bass,
lar Conf., 2003.
Tamppari et al., JGR 105(E2), 2000.
Tamppari, L. K., et al., JGR 108(E7), 2003.
Tamppari, L. K.,
Wang HQ, Ingersoll AP, JGR 107(E10), 200