09)
le
us
m/l
S ti
M
Ly
Je
a S
b S
c J
d U
e D
f I
a
Ar
Re
Re
Ac
Av
Ke
M
M
Ra
Vo
1.
tia
?
by
pr
20
la
cr
et
*
00
donn M. Carter a,?, Bruce A. Campbell a, Thomas R. Watters a, Roger J. Phillips b, Nathaniel E. Putzig b, Ali Safaeinili c,
ffrey J. Plaut c, Chris H. Okubo d, Anthony F. Egan b, Roberto Seu e, Daniela Biccari e, Roberto Orosei f
mithsonian Institution, Center for Earth and Planetary Studies, MRC 315, PO Box 37012, Washington, DC 20013-7012, USA
outhwest Research Institute, 1050 Walnut St., Suite 300, Boulder, CO 80302, USA
et Propulsion Laboratory, 4800 Oak Grove Dr., Pasadena, CA 91109, USA
.S. Geological Survey, 2255 N. Gemini Dr., Flagstaff, AZ 86001, USA
ipartimento INFOCOM, Universit? di Roma ?La Sapienza,? I-00184 Rome, Italy
stituto di Astro?sica Spaziale e Fisica Cosmica, Istituto Nazionale di Astro?sica, I-00133 Rome, Italy
r t i c l e i n f o a b s t r a c t
ticle history:
ceived 29 May 2008
vised 18 September 2008
cepted 2 October 2008
ailable online 18 November 2008
ywords:
ars
ars, surface
dar observations
lcanism
The SHARAD (shallow radar) sounding radar on the Mars Reconnaissance Orbiter detects subsurface
re?ections in the eastern and western parts of the Medusae Fossae Formation (MFF). The radar waves
penetrate up to 580 m of the MFF and detect clear subsurface interfaces in two locations: west MFF
between 150 and 155? E and east MFF between 209 and 213? E. Analysis of SHARAD radargrams
suggests that the real part of the permittivity is ?3.0, which falls within the range of permittivity values
inferred from MARSIS data for thicker parts of the MFF. The SHARAD data cannot uniquely determine the
composition of the MFF material, but the low permittivity implies that the upper few hundred meters
of the MFF material has a high porosity. One possibility is that the MFF is comprised of low-density
welded or interlocked pyroclastic deposits that are capable of sustaining the steep-sided yardangs and
ridges seen in imagery. The SHARAD surface echo power across the MFF is low relative to typical martian
plains, and completely disappears in parts of the east MFF that correspond to the radar-dark Stealth
region. These areas are extremely rough at centimeter to meter scales, and the lack of echo power is
most likely due to a combination of surface roughness and a low near-surface permittivity that reduces
the echo strength from any locally ?at regions. There is also no radar evidence for internal layering in
any of the SHARAD data for the MFF, despite the fact that tens-of-meters scale layering is apparent in
infrared and visible wavelength images of nearby areas. These interfaces may not be detected in SHARAD
data if their permittivity contrasts are low, or if the layers are discontinuous. The lack of closely spaced
internal radar re?ectors suggests that the MFF is not an equatorial analog to the current martian polar
deposits, which show clear evidence of multiple internal layers in SHARAD data.
? 2008 Elsevier Inc. All rights reserved.
Introduction
The Medusae Fossae Formation (MFF) stretches across the mar-
n equator and parts of the northern hemisphere lowlands from
140 to 240? E longitude. The surface of the MFF is dominated
wind erosion, with rough, parallel-grooved surfaces or yardangs
esent in most places (Hynek et al., 2003; Edgett and Malin,
00). It is a geologically young deposit, overlying Amazonian-aged
va ?ows in the lowlands of Elysium Planitia and Noachian-aged
atered highlands along parts of the dichotomy boundary (Hynek
al., 2003; Bradley et al., 2002; Head and Kreslavsky, 2004). Ex-
humed and buried craters are common; towards the north, the
formation is almost completely eroded to plains level (Hynek et
al., 2003; Edgett and Malin, 2000). High-resolution images show
slope streaks and small dune ?elds, which suggests that a ?ne-
grained material probably makes up most of the deposit. Several
hypotheses have been suggested to explain the origin of the MFF
deposits, including pyroclastic density currents, volcanic fall de-
posits and ignimbrites (Hynek et al., 2003; Muhleman et al., 1991;
Zimbelman et al., 1997), aeolian deposits (Tanaka, 2000), or relic
polar layered deposits (Head and Kreslavsky, 2004; Schultz and
Lutz, 1988).
The eastern parts of the MFF correspond to the radar-dark
Stealth region identi?ed in 3-cm and 13-cm ground-based radarIcarus 199 (20
Contents lists availab
Icar
www.elsevier.co
hallow radar (SHARAD) sounding observa
arsCorresponding author. Fax: +1 202 786 2612.
E-mail address: carterl@si.edu (L.M. Carter).
im
es
19-1035/$ ? see front matter ? 2008 Elsevier Inc. All rights reserved.
i:10.1016/j.icarus.2008.10.007295?302
at ScienceDirect
ocate/icarus
ons of the Medusae Fossae Formation,ages (Muhleman et al., 1991; Harmon et al., 1999). The dark-
t part of the Stealth region, which has no received echo power
29 199
ab
so
co
la
St
th
te
M
te
re
24
(E
to
ha
fr
(F
pl
20
fe
H
w
ca
M
su
H
19
si
it
jo
As
la
to
qu
Ta
be
th
as
ge
re
H
as
ev
ra
Al
of
te
su
va
sp
be
er
ve
in
SI
th
su
M
M
co
M
na
m
al
a
na
ha
in
de
to
sa
lo
m
M
ec
2.
w
in
re
di
is
re
Fo
et
ra
m
bi
ra
al
in
ha
to
un
go
an
at
no
fe
M
al
to
Fo
th
cl
gr
w
to
si
su
ra
gi
fa
th
cu
an
ri
le
lo6 L.M. Carter et al. / Icarus
ove the noise level (Muhleman et al., 1991, 1995), is located
uthwest of Tharsis. The low radar backscatter suggests some
mbination of a low surface permittivity and few cm-scale and
rger rocks (Muhleman et al., 1991; Harmon et al., 1999). The
ealth component of the MFF was inferred to be at least 5 m
ick, although the radar could potentially penetrate up to several
ns of meters in low electrical-loss material (Harmon et al., 1999;
uhleman et al., 1995). The terrain with the lowest radar backscat-
r is centered south of Olympus Mons, but the total area showing
duced radar re?ectivity stretches across the equator from 140 to
0? E, and encompasses the entire longitudinal extent of the MFF
dgett et al., 1997).
The surface properties and atmosphere of Mars are conducive
explosive volcanism. Models of martian vulcanian eruptions
ve shown that ash fall deposits may travel 25 km or more
om the vent, depending on the wind strength and direction
agents and Wilson, 1996). Convective plumes from long-lived
inian eruptions can reach heights up to 20 km (Glaze and Baloga,
02), and in this case the millimeter-sized clasts may travel a
w kilometers to 20 km depending on the winds (Wilson and
ead, 2007). Micron-sized particles may be carried farther by the
ind (Wilson and Head, 2007). Possible sites of explosive vol-
nism and ash deposits on Mars include Alba Patera (Mouginis-
ark et al., 1988), Apollinaris Patera (Robinson et al., 1993), the
mmit of Arsia Mons (Mouginis-Mark, 2002) and Tyrrhena and
adriaca Paterae (Greeley and Crown, 1990; Crown and Greeley,
93). One unresolved problem with the volcanic origin hypothe-
s for the MFF is the lack of identi?ed source regions, although
has been suggested that Tharsis-related volcanism was a ma-
r contributor (Hynek et al., 2003; Edgett and Malin, 2000).
h from explosive eruptions can also potentially settle over a
rge area, leading to low-elevation constructs that are di?cult
identify in images, and which can be easily buried by subse-
ent volcanism (Fagents and Wilson, 1996; Hynek et al., 2003;
naka et al., 2005).
If MFF is analogous to a polar layered deposit, its interior may
composed primarily of ice. Schultz and Lutz (1988) proposed
at the MFF is comprised of relic polar layered deposits formed
a result of polar wandering. Head and Kreslavsky (2004) sug-
sted that the MFF is a young, ice-rich deposit formed during
cent periods of high obliquity. Both Schultz and Lutz (1988) and
ead and Kreslavsky (2004) cite the presence of internal layering
well as the overall thickness and topography of the deposit as
idence that the MFF may be similar to martian polar layered ter-
in. However, Bradley et al. (2002) compared Mars Orbiter Laser
timeter (MOLA) data and Mars Orbiter Camera (MOC) images
the poles with those of the MFF and concluded that the two
rrains have very different characteristics, including differences in
rface slopes, layer topography, and in the heights of ridges and
lleys.
The MARSIS (Mars Advanced Radar for Subsurface and Iono-
heric Sounding) instrument on the Mars Express orbiter has also
en used to study the MFF (Watters et al., 2007). MARSIS op-
ates at frequencies between 1.3 and 5.5 MHz, has a free space
rtical resolution of 150 m, and can penetrate up to a few km
depth depending on the material (Picardi et al., 2005). MAR-
S detects basal interfaces beneath most of the MFF, including in
e regions of Gordii Dorsum, Amazonis Mensa, Eumenides Dor-
m, Lucus Planum and the far western deposits south of Elysium
ons (Watters et al., 2007). Time-delay values measured from the
ARSIS radargrams for deposits up to about 2.5 km in thickness
rrespond to a bulk real permittivity value of ?2.9 ? 0.4 for the
FF material (Watters et al., 2007). The two most likely expla-
tions for such a low permittivity are either a dry, low-density
aterial such as volcanic ash, or an ice-rich material (Watters et
ro
gr
Ra
th(2009) 295?302
., 2007). If the MFF is composed mostly of dry ash, it must have
high porosity, even at km-scale depths.
The SHARAD radar sounding instrument on the Mars Recon-
issance Orbiter (MRO) is complementary to MARSIS: SHARAD
s a higher resolution and is capable of detecting ?nely spaced
terfaces at depths of up to 1000 m, but cannot penetrate as
eply. In this paper we discuss results from a SHARAD campaign
search for subsurface interfaces across the entire Medusae Fos-
e Formation. We summarize the SHARAD data products and the
cation of subsurface interfaces across the MFF, characterize the
aterial properties of the deposit as inferred from SHARAD and
ARSIS data, and, ?nally, discuss the correlation between SHARAD
hoes and local geologic features.
Summary of SHARAD Medusae Fossae Formation observations
SHARAD operates at a center frequency of 20 MHz (? = 15 m)
ith a 10 MHz bandwidth (Seu et al., 2004, 2007a). With a nom-
al vertical resolution of 15 m in free-space, and 5?10 m vertical
solution in common geologic materials, SHARAD is capable of
scerning thin subsurface layers. The lateral resolution of SHARAD
3 to 6 km, reducible to 300 to 1000 m in the along-track di-
ction with synthetic aperture focusing (Seu et al., 2004, 2007a).
r smooth surfaces, the cross-track resolution is ?750 m (Seu
al., 2007a). On the ground, the data are correlated with the
dar chirp and spacecraft orbital information is used to deter-
ine the range relative to a topographic reference, in this case a
axial ellipsoid with dimensions equal to the polar and equatorial
dii of Mars. The SHARAD data are displayed as radargrams, with
ong-track distance on the x-axis and range-delay time increas-
g downward on the y-axis. The radargrams shown in this paper
ve been processed using a synthetic-aperture focusing algorithm
improve the signal-to-noise ratio of the subsurface re?ectors
der the rough MFF surface. Although the SHARAD processing al-
rithm will correct for the ionosphere based on the sun elevation
gle (Safaeinili et al., 2007), all the data used here were acquired
night when ionospheric distortion is minimal, and in this case
correction is applied.
The SHARAD data are sensitive to wavelength-scale topographic
atures that contribute off-nadir surface clutter to the radargrams.
OLA topographic data do not have su?cient resolution to model
l of the clutter sources seen by SHARAD, but they can be used
determine the effects of large-scale structures in the images.
r each SHARAD ground-track, we produce a simulated radargram
at can be compared with the observed radargram to identify
utter sources large enough to appear in the MOLA data. Radar-
ams projected to the surface along-track can also be compared
ith images to search for smaller clutter sources.
SHARAD has obtained radar tracks across the MFF from 140
230? E longitude, including the western portions south of Ely-
um Planitia, Apollinaris Patera, Lucus Planum, Eumenides Dor-
m, Amazonis Mensa, and Gordii Dorsum. While the MARSIS
dar detects the basal interface of the MFF along its entire lon-
tudinal extent (Watters et al., 2007), SHARAD detects subsur-
ce echoes only from the far eastern and western portions of
e deposit. The central portion of the MFF, including both Lu-
s Planum and Apollinaris Patera, has a high surface roughness
d the SHARAD radargrams are dominated by surface clutter from
dges and craters. With its longer wavelengths, MARSIS is both
ss susceptible to surface clutter and able to see deeper re?ectors.
The western parts of the MFF consist of relatively low-relief
bes that extend from the dichotomy boundary onto the sur-
unding plains. Fig. 1 shows the western MFF with SHARAD
ound tracks shown in red where there are subsurface echoes.
dargrams of some of these same orbits are shown in Fig. 2. In
is part of the MFF, SHARAD sees through low-relief hills south
Med
Fi
m
tra
M
gr
su
w
th
of
(2
N
co
N
te
in
el
to
m
th
sh
av
m
w
fe
by
M
in
ea
te
of
de
of
sh
Fi
fro
im
th
(a
Pl
far
su
re
ar
M
?a
re
tiv
po
G
la
(2
pe
anSHARAD observations of the
g. 1. SHARAD observation ground-tracks across the western part of MFF. The base
ap is THEMIS daytime infrared with color MOLA topography overlaid. The ground
cks are shown in red where SHARAD detects a subsurface interface beneath the
FF. Grey ground tracks correspond to observations shown in Fig. 2. Some available
ound tracks were omitted for clarity, but adjacent observations show the same
bsurface features. The location of the THEMIS image shown in Fig. 3 is indicated
ith a white box. Subsurface detections off the MFF in Elysium Planitia (north of
e white box) are not marked with red lines in the image.
Elysium Planitia and northeast of Aeolis Mensae. Watters et al.
007) referred to the upper part of the lobe (shown in Fig. 1) as
orth Hill, and we repeat that nomenclature in this paper to avoid
nfusion. At a height of ?580 m as measured from MOLA data,
orth Hill is the thickest part of the MFF for which SHARAD de-
cts a subsurface interface. MARSIS detects both the subsurface
terface beneath North Hill and an interface beneath a higher-
evation hill to the south that is apparently too thick for SHARAD
penetrate (Watters et al., 2007). Fig. 3 is a Mars Odyssey Ther-
al Emission Imaging System (THEMIS) daytime infrared image of
e North Hill area (Christensen et al., 2004). The THEMIS image
ows contrasts between the hills and the plains more clearly than
ailable optical imagery. The radar-transparent hills to the north,
arked as ?thin deposits? in Fig. 3, are no more than 100 m thick.
The eastern part of the Medusae Fossae Formation lies south-
est of Olympus Mons and includes large hills that are up to a
w kilometers in elevation. Subsurface interfaces detected here
SHARAD occur primarily between Gordii Dorsum and Amazonis
ensa (Fig. 4). A subset of the corresponding radargrams is shown
Fig. 5, and it is clear that the MFF deposit thickens toward the
st, from about 180 m to about 315 m. Further east, the lower in-
rface is no longer detected by the radar, presumably as a result
signal attenuation in a thicker deposit. Depth estimates were
rived from round-trip echo delay, assuming a real permittivity
3. In the eastern part of the study region, MARSIS observations
ow subsurface interfaces at ?340 and 730 m depth, again as-
3.
cousae Fossae Formation 297
g. 2. SHARAD radargrams of deposits in west MFF. The radargrams are ordered
m west (top) to east (bottom) and are shown with north to the left on each
age. Ground-tracks are indicated in Fig. 1. Subsurface interfaces occur beneath
in 50?100 m deposits (arrows marked td) and beneath the ?580 m North Hill
rrows marked nh). In some places, plains re?ectors can be seen under Elysium
anitia (arrows marked ep). A scale bar for the time-delay values is shown on the
right of each radargram.
ming a real permittivity of 3 (Watters et al., 2007). The SHARAD
?ector at ?315 m depth and the upper MARSIS re?ector likely
ise from the same subsurface interface. To the west of Amazonis
ensa, SHARAD detects a subsurface interface beneath relatively
t terrain near the dichotomy boundary (Fig. 4). In this area, the
?ectors are about 60 m below the surface (assuming a permit-
ity of 3) and are most likely boundaries between very thin de-
sits of the MFF material and the underlying plains. Northwest of
ordii Dorsum SHARAD detects dipping subsurface re?ectors simi-
r to those mapped in central Amazonis Planitia by Campbell et al.
008). These features occur underneath Late- to Mid-Amazonian
riod plains lava ?ows (labeled as AAa2s by Tanaka et al., 2005)
d do not represent an interface with the MFF materials.Physical properties of the Medusae Fossae Formation material
Sounding radar measurements can be used to estimate the
mplex permittivity (?? ? i???) of the MFF material. The real part of
29 199
Fi
Cr
ex
Fi
su
w
w
se
an
th
by
th
ne
fr
Fi
fro
tra
su
di
re
m
ra
??
w
su
be
ra
su
a
ap
po
de
an
zo
?l
N8 L.M. Carter et al. / Icarus
g. 3. THEMIS daytime infrared mosaicked image of low-relief hills in west MFF.
aters at the image center were emplaced on the underlying plains and are being
humed in the area between the thin deposits and the feature marked ?North Hill.?
g. 4. SHARAD observation ground-tracks across the eastern part of MFF with sub-
rface interface detections shown in red. The base map is THEMIS daytime infrared
ith overlaid color MOLA topography. The orbits change from black to red in places
here SHARAD detects a subsurface interface. Grey ground tracks correspond to ob-
rvations shown in Fig. 5. The locations of the HIRISE images shown in Figs. 8, 9
d 11 are indicated with white boxes.
e permittivity (i.e., the real dielectric constant) can be estimated
comparing radargrams with MOLA topography and assuming
at the MFF material lies on an constant-elevation extension of
arby plains. The permittivity of the material can be calculated
om:
pl
su
be
2.(2009) 295?302
g. 5. SHARAD radargrams of deposits in eastern MFF. The radargrams are ordered
m west (top) to east (bottom) and are shown with north to the left. Ground
cks are shown in Fig. 4. Subsurface re?ectors can be seen between Gordii Dor-
m and Amazonis Mensa (arrows marked as ga), as well as in an area near the
chotomy boundary (arrows marked db). Dipping re?ectors north of Gordii Dorsum
present contacts between geologic units in southern Amazonis Planitia (arrow
arked ap). A scale bar for the time-delay values is shown on the far right or each
dargram.
=
(
ct
2h
)2
, (1)
here h is the height relative to the surrounding plains as mea-
red from MOLA topography and t is the two-way time delay
tween the surface and subsurface echoes measured from the
dargram.
In order to illustrate the effects of different values of ??on the
bsurface pro?les, we used Eq. (1) to convert time to depth for
range of subsurface ?? values. The depth conversion causes the
parent location of subsurface echoes to migrate upward in pro-
rtion to the square root of the real permittivity. Results of these
pth conversion adjustments are shown in Fig. 6 for North Hill
d Fig. 7 for the eastern MFF between Gordii Dorsum and Ama-
nis Planitia.
In Fig. 6, a real permittivity value of 3.0 produces a depth pro-
e with a fairly ?at interface under both the thin deposit and
orth Hill. In each Fig. 6 radargram, a black line that connects the
ains adjacent to North Hill is shown for comparison with the
bsurface echo. A subsurface raised-platform structure is visible
neath the south side of North Hill. Using a real permittivity of
0, both subsurface interfaces appear to be concave upward, and
Med
Fi
va
m
on
ag
ab
re
de
fo
th
th
w
ex
pr
th
no
un
to
co
pr
od
pr
w
th
m
su
so
Pe
cl
th
no
H
va
in
pe
Fi
pe
no
clu
ar
(F
sl
se
su
of
un
pl
pe
se
th
sl
su
su
pe
2.
Fo
in
tiv
ar
a
et
si
th
ge
po
w
at
As
siSHARAD observations of the
g. 6. A portion of SHARAD observation 589803 depth-converted with different
lues of the permittivity. The full time-delay pro?le is shown in Fig. 2. A real per-
ittivity value of 3.0 produces a pro?le where the MFF deposits at North Hill lie
a substrate that gradually increases in elevation to the south (right side of im-
es). Permittivity values that are larger or smaller offset the subsurface interfaces
ove or below neighboring surface terrain. A black line is shown linking the plains
gions on either side of North Hill. For permittivity values of 2.0 or 4.0, the thin
posit to the north has a curved subsurface interface. No reference line is shown
r the thin deposit because the line would completely obscure the interface.
e interface has a bowed interface with the plains. For ?? = 4.0,
e interface beneath the thin deposit becomes slightly convex,
hich would imply that the MFF material was deposited on a pre-
isting hill. The interface below North Hill is also shallower in the
o?le than expected for continuity with the surrounding terrain.
Fig. 6 demonstrates that the initial assumption of a ?at surface
at continues at the same elevation underneath the deposits is
t a good approximation for SHARAD. If the subsurface interface
der North Hill is at the same elevation as the edge of the plains
the north (on the left side of the images), then the ?? value
mputed from Eq. (1) is 2.0. This is an extremely low value that
oduces an unlikely pro?le where the MFF material is draped over
d-shaped depressions in the plains. Instead, an ?? value of ?3.0
oduces a reasonable geologic interface with a fairly steady up-
ard slope to the south, and with ?45 m of elevation gain from
e plains to the southern edge of North Hill. Permittivity esti-
ates using MARSIS data suffer from this same uncertainty in the
bsurface interface depth, but in the MARSIS case the dominant
urce of error may be the coarser range resolution of the radar.
rmittivity estimates from MARSIS data (Watters et al., 2007) in-
ude a one-half range bin uncertainty in depth (150 m), and so
e small elevation change inferred from the SHARAD data does
t substantially change the MARSIS permittivity results for North
ill. It is not possible to compute a more precise North Hill ??
lue from the SHARAD data because the depth of the subsurface
terface and bench feature are not well-constrained. The actual
rmittivity could be somewhat higher or lower than 3.
ta
w
20
rausae Fossae Formation 299
g. 7. A portion of SHARAD observation 519702 adjusted for varying values of the
rmittivity. See Fig. 5 for the full time-delay pro?le. The slope of the ?at plains
rth of the subsurface interface is extended with a black line. Areas of surface
tter were identi?ed through comparison with simulations and are marked with
rows in the upper ground-track.
The same technique can be used in the eastern parts of the MFF
ig. 7). In this case, the MFF material appears to lie on a regional
ope, regardless of the value selected for the permittivity. The two
gments of bright plains re?ectors to the north (left) of the sub-
rface re?ector have slightly different slopes. In Fig. 7, the slope
the plains segment nearest the subsurface re?ector is continued
der the surface with a black line. Maintaining the slope of these
ains under the deposit is best done with a permittivity ?3. For a
rmittivity value of 4.0, the subsurface interface is noticeably off-
t to a higher elevation at the boundary between the plains and
e deposit. A permittivity of 2.0 places the subsurface interface
ightly below the projected slope.
This graphical approach illustrates that a permittivity of ?3 re-
lts in good ?ts of the subsurface echoes to projections of the
rrounding plains, and we use this as an estimate of the bulk
rmittivity of the MFF as seen by SHARAD. This falls within the
9?0.4 range derived by Watters et al. (2007) using MARSIS data.
r dry materials, the real part of the permittivity is primarily
?uenced by density. Pumice, volcanic ash, and tuff have permit-
ity values between ?2.5 and 3.5, while basalts have ?? values
ound 7 to 9 (Campbell and Ulrichs, 1969). Pure water ice has
permittivity of ?3.1 (Cumming, 1952). As discussed by Watters
al. (2007), it is not possible to uniquely determine the compo-
tion of the MFF based on a derived permittivity of 3. However,
e low permittivity values inferred from the SHARAD data sug-
st that the upper few hundred meters of the MFF have a high
rosity. One possibility is that these upper layers are low-density,
elded or interlocked pyroclastic units.
As a radar wave travels through a deposit, power is lost through
tenuation and scattering, resulting in a weaker subsurface echo.
suming that surface and subsurface roughness are relatively con-
stent along the orbit track, it is possible to estimate the loss
ngent of the material based on the reduction in echo power
ith round-trip delay (e.g., Campbell et al., 2008; Watters et al.,
07). Using radargrams from across the MFF, with deposit depths
nging from about 0.5 to 2.5 km, Watters et al. (2007) infer a
30
Fi
su
co
lo
fr
us
fa
ov
ac
pl
si
low SHARAD backscatter (Fig. 8). Smoother surfaces, such as the
re
su
Fi e e
in sent
57 age
th
de
an
te
(seains, and both the surface and subsurface echo power vary con-
derably along track, even as the apparent re?ector depth remains
g. 9. (a) HIRISE image (PSP_006905_1850) centered at 4.9? N, 210.8? E and showing th
terface. Color-stretched MOLA topography is overlaid on the image, with purple repre
6401 crosses this image and is marked with a black line. The position of the sub-im
e tall yardangs in the upper left. (b) THEMIS infrared image I03313002 with a box showi
posit. The plains to the north of the deposit show up as dark in the thermal infrared i
gle of 66? , where the solar incidence angle is the angle between the Sun and the surfac
rrain has fewer yardangs and a larger concentration of big blocks. (d) HIRISE detail of ro
e Fig. 5), despite the presence of these darker layers.gion between Gordii Dorsum and Amazonis Mensa, have stronger
rface echoes. MARSIS data show similar changes in surface echo
dge of a deposit southwest of Gordii Dorsum where SHARAD detects a subsurface
ing the lowest elevation values and orange the highest. The SHARAD observation
s in (c) and (d) are marked. A large pit (black arrow) is located near the base of0 L.M. Carter et al. / Icarus 199 (2009) 295?302
g. 8. HIRISE image (TRA_000865_1905) showing the rough surface of Gordii Dor-
m and centered at 10.2? N, 211.5? E. The SHARAD surface echo disappears almost
mpletely over this extremely rough terrain.
ss value of 0.0048 ? 0.0024 dB/m (tan ? = 0.002?0.006) at a
equency of 4 MHz. Although this technique can theoretically be
ed to derive loss tangents from SHARAD data as well, the inter-
ces present in the MFF are often diffuse (i.e., the power is spread
er several SHARAD range bins). In addition, the surface echo
ross the MFF deposit is reduced signi?cantly from that of the
fairly constant. These phenomena generate a large amount of scat-
ter in a comparison of echo power versus round-trip time delay.
THEMIS and High-Resolution Imaging Science Experiment (HIRISE)
images, as well as the variations in SHARAD surface backscatter,
reveal that the tens-of-meter-scale surface roughness changes sig-
ni?cantly across the MFF, adding an unquanti?able uncertainty to
loss tangent measurements derived using radargrams from across
the longitudinal extent of the deposit. The MARSIS instrument,
with a longer wavelength, is less susceptible to this type of surface
clutter. Consequently, the MARSIS radargrams show less variabil-
ity in surface and subsurface echo power than SHARAD tracks
over the same area of the MFF. Although the SHARAD data for
the MFF cannot produce a statistically robust estimate of the loss
tangent, it is clear from visual inspection of the radargrams that
these losses are quite low. For the very thin deposits, such as
the low-elevation hills in the west MFF, the subsurface interface
generally returns only slightly less echo power than the surface in-
terface.
4. SHARAD echoes and local geology of the Medusae Fossae
Formation
The SHARAD radargrams provide insight into the nature of the
MFF, particularly when compared with visible and infrared im-
agery. First, the MFF material is easily penetrable by radar and is
rough at the 15-m wavelength scale of SHARAD. The radar echo
almost completely disappears across the roughest areas, for ex-
ample, the northern tip of Gordii Dorsum (Fig. 5). HIRISE images
across Gordii Dorsum con?rm high surface roughness in areas ofng the location of the HIRISE image in (a). SHARAD sees through the light-toned
mage. The THEMIS image was acquired at 16:38 local time at a solar incidence
e normal. (c) Towards the north and away from the higher elevation deposit, the
cks eroding from the friable material. SHARAD does not detect internal layering
Med
Fi
Slo
slo
st
in
su
Th
of
po
w
du
ra
sh
te
SH
Fo
et
pl
tw
lik
fo
po
G
id
Pl
th
th
SH
M
te
su
de
sh
th
ce
de
pl
st
m
no
ca
of
be
Th
an
Fi
bl
m
to
se
gr
?a
lay
ic
be
co
m
be
in
la
m
20
H
of
in
to
la
cr
in
th
M
ol
re
th
no
so
MSHARAD observations of the
g. 10. High-resolution view of the depression near the large yardang in Fig. 9.
pe streaks can be seen descending into the pit and are a useful way to judge the
pe directions. The pit is approximately 168 m wide and 250 m long.
rength across the roughest surfaces like Gordii Dorsum, suggest-
g that surface roughness may play some role in the decreased
rface echo power even at very long (50?230 m) wavelengths.
e radar-dark behavior of the Stealth area across a wide range
wavelengths is most likely the result of a combination of radar
wer loss through transmission into a low-permittivity material
ith no strong internal re?ectors and the loss of re?ected power
e to scattering from surfaces that are rough at wavelength scales
nging from 3 cm through >100 m. The low backscatter at the
orter wavelengths suggests that in most places, the top centime-
rs of the deposit have few internal re?ectors.
There is no clear evidence of internal layering in any of the
ARAD radargrams collected to date across the Medusae Fossae
rmation. In the western part of the MFF, the thin deposits pen-
rated by SHARAD appear to lie atop an extension of the volcanic
ains of Elysium Planitia, and the strong permittivity contrast be-
een low-density MFF material and dense volcanic material most
ely creates the bright subsurface re?ection. There is no evidence
r ?ne-scale internal layering within the ?580 m North Hill de-
sit. In the east, the situation is more complex. The area between
ordii Dorsum and Amazonis Mensa consists of stepped layers, ev-
ent in MOLA data, that decrease in elevation towards Amazonis
anitia to the northwest with a slope of ?0.06? . SHARAD sees
rough only the thinnest part of these layers, which is closest to
e plains. Fig. 9a is a HIRISE image of the boundary region where
ARAD is ?rst able to detect an interface beneath this part of the
FF. A THEMIS infrared image (Fig. 9b) shows that SHARAD de-
cts the subsurface interface as the radar moves from a darker
rface onto a lighter one that may represent a dustier part of the
posit. This area corresponds to the SHARAD observation 576401
own in Fig. 5. In the northwestern part of the HIRISE image,
e terrain is partially covered by dark blocks (Fig. 9c). Near the
nter of the image, the ground track crosses onto a low-relief
posit that is easily penetrated by the SHARAD radar. In several
aces across this deposit, dark blocks can be seen eroding from
eep-sided terrain (Fig. 9d); as the material erodes, these blocks
ay be further exposed to form a surface more like that to the
rth.
Fig. 9 also shows an unusual oblong depression (Fig. 10) lo-
ted on the southeast end of a large yardang, just on the edge
the higher-elevation infrared-bright deposit. Blocky material can
seen cascading down the walls of the pit closest to the yardang.
e depression does not appear to be an impact structure. If it is
erosional feature, its size would require the removal of a signif-
SH
al
ge
deusae Fossae Formation 301
g. 11. HIRISE image (PSP_006839_1910) centered at 11.0? N and 211.7? E, showing
ocky debris shedding down a slope in eastern MFF. Three layers of more resistant
aterial can be seen along the cliff face in the vicinity of the debris falls. Farther
the north, in the upper-middle part of the image, multiple ?ne layers can be
en in the cliff face. Slope streaks cover the steep cliff face, highlighting the ?ne-
ained nature of the MFF material. Although the surface to the west is relatively
t compared to other areas of the MFF, SHARAD does not detect these internal
ers.
ant amount of material from depth. Instead, the depression may
a pit crater or a relic vent structure, particularly if the MFF is
mposed of volcanic deposits. If the pit is a volcanic vent, then
ost of the original surrounding volcanic structure has probably
en removed by the wind, perhaps leaving only the portion that
cludes the large yardang.
Despite the lack of evidence in the SHARAD data for internal
yering within the MFF, evidence of such layering has been docu-
ented in MOC, THEMIS, and HIRISE images (e.g., Bradley et al.,
02; Hynek et al., 2003; Zimbelman et al., 1997). Fig. 11 is a
IRISE image of an area north of Gordii Dorsum. In the middle
the image, three dark layers a few meters apart are eroding
to blocks that form rock slides along the steep slopes. Farther
the north, multiple ?ne layers can be seen in the cliffs. The
yers are ?at and parallel, with no evidence of discontinuities or
oss-bedding. Slope streaks are also prevalent in this area, show-
g the ?ne-grained nature of the surface materials across most of
e area.
There are several reasons why SHARAD might not detect the
FF layers seen in images. Layers could be spaced below the res-
ution limit of the radar, they may be discontinuous over large
gions, or, perhaps most likely, the permittivity contrast between
e ?ne-grained material and the block-forming dark material may
t be su?cient to produce a strong re?ection. Whatever the rea-
ns for the lack of internal interfaces, it is apparent that the
FF lacks the same type of ?ne-scale permittivity layering that
ARAD easily detects within the martian polar caps (e.g., Seu et
., 2007b; Phillips et al., 2008). The lack of internal re?ectors sug-
sts that the MFF is not an analog to the current martian polar
posits.
302 L.M. Carter et al. / Icarus 199 (2009) 295?302
Acknowledgments
We thank the SHARAD Operations Center team, including
Emanuele Giacomoni, Federica Russo, Marco Cutigni, Oreste Fuga,
and Riccardo Mecozzi for their assistance with targeting, calibra-
tion, and data processing. We thank the HIRISE and MRO engi-
neering and operations teams for their work in designing, building,
and operating the equipment. We are grateful to Fabrizio Bernar-
dini for assistance with SHARAD data processing. Lionel Wilson
and an anonymous reviewer provided helpful comments. The Shal-
low Subsurface Radar (SHARAD) was provided by the Italian Space
Agency through a contract with Thales Alenia Space Italia, and it
is operated by the INFOCOM Department, University of Rome ?La
Sa
Pa
Re
Br
Ca
Ca
Ch
Cr
Cu
Ed
Ed
Fa
Gl
Gr
Ha
He
Hy
M
Mouginis-Mark, P.J., Wilson, L., Zimbelman, J.R., 1988. Polygenic eruptions on Alba
Patera, Mars. Bull. Volcanol. 50, 361?379.
Muhleman, D.O., Butler, B.J., Grossman, A.W., Slade, M.A., 1991. Radar images of
Mars. Science 253, 1508?1513.
Muhleman, D.O., Grossman, A.W., Butler, B.J., 1995. Radar investigations of Mars,
Mercury, and Titan. Annu. Rev. Earth Planet. Sci. 23, 337?374.
Phillips, R.J., Zuber, M.T., Smrekar, S.E., Mellon, M.T., Head, J.W., Tanaka, K.L., Putzig,
N.E., Milkovich, S.M., Campbell, B.A., Plaut, J.J., Safaeinili, A., Seu, R., Biccari, D.,
Carter, L.M., Picardi, G., Orosei, R., Mohit, P.S., Heggy, E., Zurek, R.W., Egan, A.F.,
Giacomoni, E., Russo, F., Cutigni, M., Pettinelli, E., Holt, J.W., Leuschen, C.J., Mari-
nangeli, L., 2008. Mars North Polar deposits: Stratigraphy, age, and geodynamical
response. Science 320, 1182?1185.
Picardi, G., Plaut, J.J., Biccari, D., Bombaci, O., Calabrese, D., Cartacci, M., Cicchetti, A.,
Clifford, S.M., Edenhofer, P., Farrell, W.M., Federico, C., Frigeri, A., Gurnett, D.A.,
Hagfors, T., Heggy, E., Herique, A., Huff, R.L., Ivanov, A.B., Johnson, W.T.K., Jordan,
Ro
Sa
Sc
Se
Se
Se
Ta
Ta
W
W
Zipienza.? This work was partially supported through a NASA MRO
rticipating Scientist grant to L.M. Carter.
ferences
adley, B.A., Sakimoto, S.E.H., Frey, H., Zimbelman, J.R., 2002. Medusae Fossae For-
mation: New perspectives from Mars Global Surveyor. J. Geophys. Res. 107 (E8),
doi:10.1029/2001JE001537. 5058.
mpbell, B.A., Carter, L.M., Phillips, R.J., Putzig, N.E., Plaut, J.J., Safaeinili, A., Seu,
R., Biccari, D., Orosei, R., 2008. SHARAD radar sounding of Amazonis Planitia.
J. Geophys. Res., doi:10.1029/2008JE003177, in press.
mpbell, M., Ulrichs, J., 1969. Electrical properties of rocks and their signi?cance
for Lunar radar observations. J. Geophys. Res. 74, 5867?5881.
ristensen, P.R., Jakowsky, B.M., Kieffer, H.H., Malin, M.C., McSween, H.Y., Nealson,
K., Mehall, G.L., Silverman, S.H., Ferry, S., Caplinger, M., Ravine, M., 2004. The
Thermal Emission Imaging System (THEMIS) for the Mars 2001 Odyssey Mis-
sion. Space Sci. Rev. 110, 85?130.
own, D.A., Greeley, R., 1993. Volcanic geology of Hadriaca Patera and the Eastern
Hellas region on Mars. J. Geophys. Res. 98, 3431?3451.
mming, W.A., 1952. The dielectric properties of ice and snow at 3.2 centimeters.
J. Appl. Phys. 23, 768?773.
gett, K.S., Malin, M.C., 2000. A meter-scale view of the Mars radar ?Stealth? South-
west Tharsis. Lunar Planet. Sci. 31. Abstract #1065.
gett, K.S., Butler, B.J., Zimbelman, J.R., Hamilton, V.E., 1997. Geologic context of
the Mars radar ?Stealth? region in southwestern Tharsis. J. Geophys. Res. 102,
21545?21567.
gents, S.A., Wilson, L., 1996. Numerical modeling of ejecta dispersal from transient
volcanic explosions on Mars. Icarus 123, 284?295.
aze, L.S., Baloga, S.M., 2002. Volcanic plume heights on Mars: Limits of validity for
convective models. J. Geophys. Res. 107 (E10), doi:10.1029/2001JE001830. 5086.
eeley, R., Crown, D.A., 1990. Volcanic geology of Tyrrhena Patera, Mars. J. Geophys.
Res. 95, 7133?7149.
rmon, J.K., Arvidson, R.E., Guinness, E.A., Campbell, B.A., Slade, M.A., 1999. Mars
mapping with delay-Doppler radar. J. Geophys. Res. 104, 14065?14090.
ad, J.W., Kreslavsky, M., 2004. Medusae Fossae Formation: Ice-rich airborne dust
deposited during periods of high obliquity? Lunar Planet. Sci. 35. Abstract
#1635.
nek, B.M., Phillips, R.J., Arvidson, R.E., 2003. Explosive volcanism in the Tharsis
region: Global evidence in the martian geologic record. J. Geophys. Res. 108 (E9),
doi:10.1029/2003JE002062. 5111.
ouginis-Mark, P.J., 2002. Prodigious ash deposits near the summit of Arsia Mons
volcano, Mars. Geophys. Res. Lett. 29 (16), doi:10.1029/2002GL015296. 1768.R.L., Kirchner, D.L., Kofman, W., Leuschen, C.J., Nielsen, E., Orosei, R., Pettinelli,
E., Phillips, R.J., Plettemeier, D., Safaeinili, A., Seu, R., Stofan, E.R., Vannaroni, G.,
Watters, T.R., Zampolini, E., 2005. Radar soundings of the subsurface of Mars.
Science 310, 1925?1928.
binson, M.S., Mouginis-Mark, P.J., Zimbelman, J.R., Wu, S.S.C., Ablin, K.K.,
Howington-Kraus, A.E., 1993. Chronology, eruption duration, and atmospheric
contribution of the martian volcano Apollinaris Patera. Icarus 104, 301?323.
faeinili, A., Kofman, W., Mouginot, J., Gim, Y., Herique, A., Ivanov, A.B., Plaut,
J.J., Picardi, G., 2007. Estimation of the total electron content of the martian
ionosphere using radar sounder echoes. Geophys. Res. Lett. 34, doi:10.1029/
2007GL032154. L23204.
hultz, P., Lutz, A.B., 1988. Polar wandering of Mars. Icarus 73, 91?141.
u, R., Biccari, D., Orosei, R., Lorenzoni, L.V., Phillips, R.J., Marinangeli, L., Picardi, G.,
Masdea, A., Zampolini, E., 2004. SHARAD: The MRO 2005 shallow radar. Planet.
Space Sci. 52, 157?166.
u, R., Phillips, R.J., Biccari, D., Orosei, R., Masdea, A., Picardi, G., Safaeinili, A.,
Campbell, B.A., Plaut, J.J., Marinangeli, L., Smrekar, S.E., Nunes, D.C., 2007a.
SHARAD sounding radar on the Mars Reconnaissance Orbiter. J. Geophys.
Res. 112, doi:10.1029/2006JE002745. E05S05.
u, R., Phillips, R.J., Alberti, G., Biccari, D., Bonaventura, F., Bortone, M., Calabrese,
D., Campbell, B.A., Cartacci, M., Carter, L.M., Catallo, C., Croce, A., Croci, R.,
Cutigni, M., Di Placido, A., Dinardo, S., Federico, C., Flamini, E., Fois, F., Frigeri, A.,
Fuga, O., Giacomoni, E., Gim, Y., Guel?, M., Holt, J.W., Kofman, W., Leuschen, C.J.,
Marinangeli, L., Marras, P., Masdea, A., Mattei, S., Mecozzi, R., Milkovich, S.M.,
Morlupi, A., Mouginot, J., Orosei, R., Papa, C., Patern?, T., Persi del Marmo, P.,
Pettinelli, E., Pica, G., Picardi, G., Plaut, J.J., Provenziani, M., Putzig, N.E., Russo,
F., Safaeinili, A., Salzillo, G., Santovito, M.R., Smrekar, S.E., Tattarletti, B., Vicari,
D., 2007b. Accumulation and erosion of Mars? South Polar layered deposits. Sci-
ence 317, 1715?1718.
naka, K.L., 2000. Dust and ice deposition in the martian record. Icarus 144, 254?
266.
naka, K.L., Skinner, J.A., Hare, T.M., 2005. Geologic map of the northern plains of
Mars. US Geol. Surv. Sci. Inv. Map 2888.
atters, T.R., Campbell, B.A., Carter, L.M., Leuschen, C.J., Plaut, J.J., Picardi, G., Orosei,
R., Safaeinili, A., Clifford, S.M., Farrell, W.M., Ivanov, A.B., Phillips, R.J., Stofan,
E.R., 2007. Radar sounding of the Medusae Fossae Formation, Mars: Equatorial
ice or dry, low-density deposits? Science 318, 1125?1128.
ilson, L., Head, J.W., 2007. Explosive volcanic eruptions on Mars: Tephra and ac-
cretionary Iapilli formation, dispersal and recognition in the geologic record. J.
Volcanol. Geotherm. Res. 163, 83?97.
mbelman, J.R., Crown, D.A., Grant, J.A., Hooper, D.M., 1997. The Medusae Fossae
Formation, Amazonis Planitia, Mars: Evaluation of proposed hypothesis of ori-
gin. Lunar Planet. Sci. 28. Abstract #1482.