Mars: Surface and Interior 327
and tracks made by dust devils are visible on many high-
resolution images taken from orbit. Dust can be seen draped
over rocks in most lander images. Crater tails caused by eo-
lian deposition or erosion in the lee of craters are common.
Dunes are visible in almost all orbiter images with resolu-
tions of a few m/pixel or better, and in some areas, such those
as around the North Pole, dunes cover vast areas. Given all
this evidence, it is somewhat surprising that wind erosion
is not more widespread. The wind appears to mostly move
loose material around the surface. Additions to the inven-
tory of loose material by erosion of primary rocks must be
proceeding extremely slowly.
Though the effects of wind erosion in most places are
trivial, locally the effects may be substantial. This is par-
ticularly true where friable deposits are at the surface. In
southern Amazonis and south of Elysium Planitia, thick,
easily erodible deposits cover the plains/upland boundary.
Eroded into these deposits are arrays of curvilinear, paral-
lel grooves that resemble terrestrial wind-cut grooves called
yardangs. Wherever such wind erosion is observed, other
evidence indicates that what is being eroded is a deposit
that blankets the bedrock. Erosion of bedrock units such as
lava flows is minute. Wind may be ineffective as an erosion
agent because of the lack of abrasive debris for the wind to
move. On Earth, quartz sand is an effective erosion agent,
but quartz sand is rare or absent on Mars. Most of the loose
material blown around by the wind appears to be ground
up basalt and its weathering products, and these materials
have little abrasive capacity.
8. Poles
During fall and winter, CO 2 condenses onto the polar re-
gions to form a seasonal cap that can extend as far equator-
ward as 40◦latitude. In summer, the CO 2 cap sublimates.
That in the north sublimates completely to expose a water
ice cap, the temperature at the pole rises from the frost
point of CO 2 (150 K) to the frost point of water (200 K),
and the amount of water vapor over the pole rises dramat-
ically. In the south, the CO 2 cap does not dissipate com-
pletely, but water ice has been detected under the seasonal
cap.
At both poles, layered deposits several kilometers thick
extend out to roughly the 80◦latitude. Individual layers
are best seen in the walls of valleys cut into the sediments,
where layering is observed at a range of scales down to
the resolution limit of our best pictures. The frequency of
impact craters on the upper surface of the deposits suggests
that the sediments are young, of the order of 10^8 years or
less.
The poles act as a cold trap for water. Any water entering
the atmosphere as a result of geologic processes such as
volcanic eruptions or floods will ultimately be frozen out
at the poles. The poles may also be a trap for dust, in that
dust can be scavenged out of the atmosphere as CO 2 freezes
onto the poles each fall and winter. The layered deposits are,
therefore, probably mixtures of dust and ice. The layering
is thought to be caused in some way by periodic changes in
the thermal regimes at the poles, induced by variations in
the planet’s orbital and rotational motions (see Section 2.1).
These cyclical motions affect temperatures at the poles, the
stability of CO 2 and H 2 O, the pressure and circulation of
the atmosphere, the incidence of dust storms and so forth;
hence, the belief that they are responsible in some way for
the observed layering.9. The View from the Surface
At the time of writing, we had successfully landed at five
locations on the martian surface: twoVikingspacecraft in
1976,Mars Pathfinderin 1997, and two rovers in 2004.
Viking 1landed on a rolling, rock-strewn plain partly cov-
ered with dunes in the Chryse basin.Viking 2landed on
a level, rocky plain in Utopia. The main goal of theViking
landers was life detection. They carried a complex array
of experiments designed to detect metabolism in different
ways and to determine what organics might be in the soil.
Neither metabolism nor organics were detected. The lack
of organics was somewhat surprising since organics should
have been there from meteorite infall. The soil, however,
turned out to be oxidizing, which probably caused decom-
position of any organics that might at one time have been
present.Pathfinderalso landed on a rock-strewn plain in
Chryse. The site is at the mouth of one of the large out-
flow channels. It was hoped that evidence of floods might
be observed there. However, the only sign of floods were
some rocks stacked on edge and terraces on nearby hills
that could have been shorelines.
The two roversSpiritandOpportunity, launched in
2003, have been far more fruitful and have provided the
first solid evidence from the surface for pooling of water
and aqueous alteration.Spiritlanded on the flat floor of
the 160-km-diameter crater Gusev. The site was chosen
because the southern wall of Gusev is breached by a large
channel called Ma’adim Vallis. Water from the channel must
at one time have pooled in Gusev, and it was hoped that the
rover would be able to sample sediments from the pos-
tulated Gusev lake. The floor of Gusev turned out to be
another rock-strewn plain. The rocks are basalts, but they
have alteration rinds with varying amounts of water-soluble
components such as S, Cl, and Br. The alteration is minor
and has been attributed to the action of acid fogs. Erosion
rates estimated from craters superimposed on the plains
indicate that the rates have been several orders of mag-
nitude less than typical terrestrial rates. These somewhat
disappointing results spurred a move to some nearby hills,
where it was hoped different materials would be found,
and indeed they were. Most of the rocks on the Columbia
Hills are very different from those on the plains. As of this
writing, six different classes of rocks had been identified