Spirit Mars Rover Mission to the Columbia Hills, Gusev Crater:
Mission overview and selected results from the Cumberland Ridge to
Home Plate
R. E. Arvidson,1 S. W. Ruff,2 R. V. Morris,3 D. W. Ming,3 L. S. Crumpler,4 A. S. Yen,5
S. W. Squyres,6 R. J. Sullivan,6 J. F. Bell III,6 N. A. Cabrol,7 B. C. Clark,8 W. H. Farrand,9
R. Gellert,10 R. Greenberger,1 J. A. Grant,11 E. A. Guinness,1 K. E. Herkenhoff,12
J. A. Hurowitz,5 J. R. Johnson,12 G. Klingelho?fer,13 K. W. Lewis,14 R. Li,15 T. J. McCoy,16
J. Moersch,17 H. Y. McSween,17 S. L. Murchie,18 M. Schmidt,15 C. Schro?der,3 A. Wang,1
S. Wiseman,1 M. B. Madsen,19 W. Goetz,20 and S. M. McLennan21
Received 12 May 2008; revised 1 July 2008; accepted 31 July 2008; published 6 November 2008.
[1] This paper summarizes the Spirit rover operations in the Columbia Hills of Gusev
Crater from sols 513 to 1476 and provides an overview of selected findings that focus on
synergistic use of the Athena Payload and comparisons to orbital data. Results include
discovery of outcrops (Voltaire) on Husband Hill that are interpreted to be altered impact
melt deposits that incorporated local materials during emplacement. Evidence for
extensive volcanic activity and aqueous alteration in the Inner Basin is also detailed,
including discovery and characterization of accretionary lapilli and formation of sulfate,
silica, and hematite-rich deposits. Use of Spirit?s data to understand the range of spectral
signatures observed over the Columbia Hills by the Mars Reconnaissance Orbiter?s
Compact Reconnaissance Imaging Spectrometer (CRISM) hyperspectral imager
(0.4?4 mm) is summarized. We show that CRISM spectra are controlled by the proportion
of ferric-rich dust to ferrous-bearing igneous minerals exposed in ripples and other wind-
blown deposits. The evidence for aqueous alteration derived from Spirit?s data is
associated with outcrops that are too small to be detected from orbital observations or with
materials exposed from the shallow subsurface during rover activities. Although orbital
observations show many other locations on Mars with evidence for minerals formed or
altered in an aqueous environment, Spirit?s data imply that the older crust of Mars has
been altered even more extensively than evident from orbital data. This result greatly
increases the potential that the surface or shallow subsurface was once a habitable regime.
Citation: Arvidson, R. E., et al. (2008), Spirit Mars Rover Mission to the Columbia Hills, Gusev Crater: Mission overview and
selected results from the Cumberland Ridge to Home Plate, J. Geophys. Res., 113, E12S33, doi:10.1029/2008JE003183.
1. Introduction
[2] The Mars Exploration Rover, Spirit, touched down on
the volcanic plains of Gusev Crater on 4 January 2004.
During its first 156 sols Spirit conducted traverses and made
measurements on the olivine-bearing basaltic plains that
JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 113, E12S33, doi:10.1029/2008JE003183, 2008
Click
Here
for
Full
Article
1Department of Earth and Planetary Sciences, Washington University,
St. Louis, Missouri, USA.
2School of Earth and Space Exploration, Arizona State University,
Tempe, Arizona, USA.
3Johnson Space Center, NASA, Houston, Texas, USA.
4New Mexico Museum of Natural History and Science, Albuquerque,
New Mexico, USA.
5Jet Propulsion Laboratory, California Institute of Technology,
Pasadena, California, USA.
6Department of Astronomy, Cornell University, Ithaca, New York, USA.
7NASA Ames/SETI Institute, Moffett Field, California, USA.
8Lockheed Martin Corporation, Littleton, Colorado, USA.
9Space Science Institute, Boulder, Colorado, USA.
10Department of Physics, University of Guelph, Guelph, Ontario,
Canada.
Copyright 2008 by the American Geophysical Union.
0148-0227/08/2008JE003183$09.00
E12S33
11Center for Earth and Planetary Studies, National Air and Space
Museum, Smithsonian Institution, Washington, District of Columbia, USA.
12U. S. Geological Survey, Flagstaff, Arizona, USA.
13Institut fu?r Anorganische und Analytische Chemie, Johannes Gutenberg-
Universita?t, Mainz, Germany.
14California Institute of Technology, Pasadena, California, USA.
15Department of Civil and Environmental Engineering and Geodetic
Science, Ohio State University, Columbus, Ohio, USA.
16Department of Mineral Sciences, Smithsonian Institution, Washington,
District of Columbia, USA.
17Department of Earth and Planetary Sciences, University of Tennessee,
Knoxville, Tennessee, USA.
18Applied Physics Laboratory, Johns Hopkins University, Laurel,
Maryland, USA.
19Niels Bohr Institute, University of Copenhagen, Copenhagen, Denmark.
20Max Planck Institute for Sonnensystemforschung, Katlenburg-
Lindau, Germany.
21Department of Geosciences, State University of New York, Stony
Brook, New York, USA.
1 of 35
dominate the floor of the Crater [Arvidson et al., 2006a]. It
then drove onto the older Columbia Hills and has been
traversing and making measurements of the terrains,
rocks, and soils within the Hills since then (Figures 1?3
and A1?A8, Table 2). Results for the mission that include
analyses of plains and initial Columbia Hills data (including
West Spur and a portion of Husband Hill, up to and
including measurements on the float rock, Backstay, sol
512, on the Cumberland Ridge), were reported in numerous
papers in two Journal of Geophysical Research 2006
Special Issues. In this paper, a mission narrative is provided
for the period when the rover left Backstay until it parked
on the northern flank of Home Plate (sol 1447) to wait out
its third Martian winter and associated low solar energy
conditions (Figure 3). Selected scientific results are pre-
sented in this paper that illustrate the synergistic use of the
rover?s Athena Science Payload (Table 1) [Squyres et al.,
2003] and that focus on understanding the geologic evolu-
tion of the Columbia Hills and implications for the role of
water in modifying crustal materials. This paper comple-
ments papers that provide detailed findings about the
geology, chemistry, and mineralogy of Husband Hill and
the Inner Basin that are included in the third and fourth
Journal of Geophysical Research Special Issues scheduled
for publication in 2008.
[3] Since publication of the initial Journal of Geophysical
Research MER Special Issues in 2006, the Mars Recon-
naissance Orbiter (MRO) has begun operations and the
Columbia Hills have been imaged with the Compact Re-
connaissance Imaging Spectrometer for Mars (CRISM)
[Murchie et al., 2007], Context Imager (CTX) [Malin et
al., 2007] and High Resolution Imaging Science Experi-
ment (HiRISE) instruments [McEwen et al., 2007]. Analy-
ses of selected MRO data are included in this paper to help
define the regional-scale geomorphic and geologic contexts
for Spirit?s observations and to extend the type of informa-
tion that can be derived from orbit down to the fine details
that can be observed from a mobile surface platform.
Figure 1. CRISM-based false color infrared composite for the Columbia Hills (
4.5 km wide at bottom
margin) and surrounding cratered plains in Gusev Crater, with key features labeled. Spirit rover traverse
locations from the landing site to the Low Ridge winter campaign site (where Spirit spent its second
winter) located to the southeast of Home Plate are overlain as red line. Box denotes location for which a
HiRISE image subset is shown in Figure 2. CRISM data with 18 m/pixel spatial from FRT00003192_07
using band 1098 nm for blue, 1518 for green, and 2528 nm for red are used in the composite. CTX image
data with 6 m/pixel (frame PSP_001513_1654_XI_14S184W_061122) were used as the intensity in the
color image to sharpen fine detail. North is to the top of this image. The striping is an artifact. Mars
equirectangular projection.
E12S33 ARVIDSON ET AL.: MER SPIRIT OVERVIEW AND SELECTED RESULTS
2 of 35
E12S33
Finally, implications for the extent of alteration of crustal
materials on Mars by aqueous fluids are summarized on the
basis of combined results from surface and orbital measure-
ments for the Columbia Hills. Place names used in this
paper are informal, with the exception of Meridiani Planum,
Gusev Crater, and the Kau Desert.
2. Mission Narrative
[4] The operational approach for driving, approaching
targets, and making measurements with Spirit (and Oppor-
tunity) and its Athena Payload (Table 1) has been detailed in
previous papers [Arvidson et al., 2006a; Squyres et al., 2006]
and will not be repeated here. Rather the focus is on a mission
narrative, using the traverse locations overlain on MRO
image data (Figures 1?2) to show the local and regional
geomorphic context for the rover?s observations. Table 2
provides a detailed summary of the traverse andmeasurement
campaigns for Spirit for the period relevant to this paper,
Figure 3 provides a graphical view of operations as a function
of time with Martian southern hemisphere seasons delineated,
and Appendix A provides a detailed set of traverse and
experiment site maps superimposed onto a HiRISE-based
map. As shown in Figures 1?2 and Appendix A, Spirit?s
traverses for the time period covered in this paper took place
in the Columbia Hills, specifically Husband Hill and the
Inner Basin. The Columbia Hills are a triangularly shaped
complex of hills and intervening valleys that is embayed
Figure 2a. HiRISE subframe covering Spirit?s traverse and experiment sites (yellow line) in the
Columbia Hills. Boxes show traverse regions examined in detail in subsequent Figures 2b?2d. Plan view
generated using PSP_001513_1655_red image. Image covers 
1000 m in width.
E12S33 ARVIDSON ET AL.: MER SPIRIT OVERVIEW AND SELECTED RESULTS
3 of 35
E12S33
Figure 2b. HiRISE subframe covering the western and southwestern portions of Husband Hill and the
Cumberland Ridge and a portion of the Tennessee Valley. Traverses are shown in yellow, along with
locations for each sol.
Figure 2c. HiRISE subframe covering a portion of the southern slope of Husband Hill, including the
dark El Dorado ripple field, bright ripples extending to the southeast, and bright outcrops to the east and
northeast of the El Dorado feature.
E12S33 ARVIDSON ET AL.: MER SPIRIT OVERVIEW AND SELECTED RESULTS
4 of 35
E12S33
by younger olivine-bearing volcanic plains [McCoy et al.,
2008]. From West Spur, where Spirit entered the Columbia
Hills, the distance to the eastern boundary with the plains is

1 km (Figure 1). Husband Hill, which was investigated in
detail by Spirit, is
0.5 km wide (E?W) and has a maximum
elevation above the plains of 
80 m (Figures 2a?2c). The
Inner Basin is located to the south of Husband Hill and the
northwest of McCool Hill, and is dominated by a series of
interconnected low hills that include Home Plate, an oval
feature 
80 m wide that rises a few meters above the
surrounding terrain (Figures 2a?2d). Mitcheltree Ridge
and Low Ridge, oval to elongate hills that are located to
the east and southeast of Home Plate, respectively, were also
investigated by Spirit in some detail (Figures 2a?2d).
[5] Spirit left the basaltic float rock, Backstay, on sol 513
and continued its ascent to the summit of Husband Hill,
traversing along the western slopes and cutting at low
angles across topographic contours (to avoid excessive
wheel slippage) to reach the western portion of the summit
on sol 580. Immediately after leaving Backstay the outcrop
Methuselah was examined using the payload instruments.
The outcrop Independence was then examined in detail,
including use of the right front wheel to scuff the bedrock
and remove dust and coatings [Clark et al., 2007]. Inde-
pendence is a new rock type (relative to prior discoveries,
[Squyres et al., 2006]) characterized by the lowest FeO
content (3.8% weight) of any rock measured on Mars [Ming
et al., 2008]. Spirit then encountered a suite of layered
outcrops (Voltaire) and spent approximately 20 sols collect-
ing in situ measurements for a series of targets on the
outcrops while the remote sensing instruments observed
dozens of targets within and beyond the outcrops (Table 3).
The Voltaire experiments revealed a subclass of the Inde-
pendence class of rocks (Assemblee) characterized by a
high Cr2O3 content (2.9%) and a new class, Descartes, that
is interpreted to be altered impact material (detailed in
section 3 and by Ming et al. [2008]). The rover then
continued its ascent to the top of Husband Hill, reached
the summit and began a remote sensing campaign of the
surrounding terrains from its high vantage point, including
looking into the Inner Basin for safe paths for its descent to
Home Plate. It then investigated an aeolian drift soil deposit
(Lambert) before conducting observations on the float rock,
Irvine, a subalkaline basalt rock class [McSween et al.,
2006; Ming et al., 2008]. The Cliffhanger aeolian ripples at
the head of the Tennessee Valley were then investigated,
followed by the rock outcrop, Hillary (similar to the
Watchtower rock class found on the northwestern slopes
of Husband Hill [Squyres et al., 2006; Ming et al., 2008]),
located on the eastern side of the summit.
[6] The next phase of the mission was the descent into the
Inner Basin by way of Haskin Ridge (Figure 2b). The
Kansas Outcrop was characterized (similar to Hillary [Ming
et al., 2008]), followed by a suite of measurements for
olivine-rich outcrops (new Algonquin class rocks: Larry?s
Bench, Seminole, Algonquin, Comanche [Morris et al.,
2008; Ming et al., 2008]) before approaching the dark ripple
field, El Dorado, located on the southeastern slopes of
Husband Hill. The rover traversed to the edge of the ripple
field and conducted remote sensing and in situ measure-
ments of these materials before continuing its drive to Home
Plate. During one of its drives, after leaving El Dorado,
Spirit encountered difficulty in reaching a waypoint because
of wheel slippage. The wheels churned up shallow subsur-
face weakly bound rock or soil material that were found to
Figure 2d. HiRISE subframe covering the Inner Basin, centered on Home Plate. Key features are
labeled. Second and third winters correspond to locations where Spirit parked for these periods.
E12S33 ARVIDSON ET AL.: MER SPIRIT OVERVIEW AND SELECTED RESULTS
5 of 35
E12S33
be enriched in hydrated ferric sulfate minerals (Arad depos-
its) [Johnson et al., 2007; Yen et al., 2008; Wang et al.,
2008]. Detailed measurements were made of these materials
before the rover continued its drives toward Home Plate.
[7] Home Plate, a partially eroded volcaniclastic con-
struct [Squyres et al., 2007], was reached on sol 746 and
a detailed measurement campaign was begun on the layered
outcrop materials that showed fining upward sequences,
along with extensive cross bedding (Barnhill, Posey (float),
and Cool Papa Bell outcrops that define the Barnhill class of
volcaniclastic rocks [Ming et al., 2008]). A float rock,
Fuzzy Smith, was encountered on the northeastern edge of
Home Plate, and measurements showed it to be a new rock
class highly enriched in silica, titanium, and perhaps iron
sulfide minerals relative to other rock measurements by
Spirit [Squyres et al., 2007]. The rover then drove south
through the Eastern Valley between the eastern edge of
Home Plate and the western side of Mitcheltree Ridge. The
intent was to head for the northern slopes of McCool Hill to
find a suite of measurement locations for the rover to spend
its second Mars winter. The need to park over the winter
season was a consequence of the fact that Spirit is located at
almost 15
S latitude and during the winter solstice the sun is
directly overhead at 25
N latitude. Combined with accu-
mulating dust on the solar panels, this situation called for a
northerly vehicle tilt to maximize receipt of sunlight to
maintain a reasonable battery charge. During its drive to
McCool Hill, Spirit encountered another region with exten-
sive wheel slippage associated with a gentle rise and terrace
and excavated another high sulfate soil material (Tyrone). It
Figure 3. Spirit mission timeline from sol 500 to 1500. RAT brush operations are shown as circles
(RAT grind was inoperative during the period shown), wheel-based scuffs to expose rock substrate
(Independence) and soils are shown as squares, and soil targets exposed by wheel motions during drives
are shown as triangles. Southern hemisphere seasons are shown in color coded form. Dashed lines show
periods when Spirit spent its second and third winters. Italicized text indicates winter campaign activities
that were not RAT brush, scuff, or disturbed experiments. This figure is meant to complement more
detailed listings of activities provided in Table 2.
E12S33 ARVIDSON ET AL.: MER SPIRIT OVERVIEW AND SELECTED RESULTS
6 of 35
E12S33
became clear that traversing to McCool Hill would be
difficult indeed, particularly when it became evident that
the front right wheel drive motor had failed. Thus the
decision was made to drive expeditiously to a nearby
north-facing slope to spend the winter. The vehicle was
commanded to drive to the northeast slope of Low Ridge
and achieved an 8
 northerly tilt, a value sufficient to
survive its second winter season.
[8] Spirit spent sols 805 to 1037 at the Low Ridge site
during its winter campaign phase and conducted a suite of
experiments focused on long duration Alpha Particle X-Ray
Spectrometer (APXS) and Mo?ssbauer Spectrometer (MB)
observations of soils and rocks within the Instrument
Deployment Device (IDD) work volume, acquisition of a
13 filter high fidelity ??McMurdo Panorama?? using the
Pancam imaging system (see Table 1 for instrument descrip-
tions), photometry studies, monitoring surface and atmo-
spheric targets for temporal changes using the various
imaging systems (Table 1), and obtaining hundreds of
emission spectra of the atmosphere and surface using the
Mini-Thermal Emission Spectrometer (Mini-TES). The
rock experiments included multiple in situ measurements
of the platy buff-colored outcrops typified by Halley (rock
class characterized by enrichment in hematite relative to
other rocks [Morris et al., 2008; Ming et al., 2008]) and an
attempt to use the Rock Abrasion Tool (RAT) brush to
progressively bore into deeper and deeper soil horizons with
measurements made at each horizon (Progress soil experi-
ments). The inoperative right front wheel was dragged from
Tyrone back to the winter campaign site as the rover drove
backward. In a serendipitous turn of events, sulfate-rich
Tyrone soil deposits excavated while the vehicle was in the
Tyrone area were caught within the right front wheel
cowling and released along the 
40 m drive to Low Ridge,
including deposits within the winter campaign work vol-
ume. Long duration measurements were made on these
materials (Berkner Island). Finally, long duration magnet
measurements were also made during the winter campaign
[Madsen et al., 2008].
[9] Leaving the winter campaign site was done with care
toward the end of the winter season, first with a short bump
to finely layered granular outcrops within Low Ridge
(Graham Land, King George Island target) and then to a
vesicular basalt float rock (Esperanza) also located on Low
Ridge. The Esperanza measurements were cut short because
of rising atmospheric dust opacity and thus decreasing solar
energy on the panels. Spirit was sent on a short drive to a
more northerly tilted terrain that also placed the vehicle for
in situ measurements on the Troll outcrops (Montalva and
Riquelme targets, a new rock class defined by high hematite
and K2O contents [Morris et al., 2008; Ming et al., 2008]).
Next Spirit was directed to drive back to within 10 m of the
Tyrone disturbed soil deposits to acquire remote sensing and
in situ data to further characterize the deposits and to search
Table 1. Athena Payload and Engineering Camera Descriptions
Instrument Key Parameters
Mast Mounted
Panoramic Camera
(Pancam)
Mulitspectral imager (0.4?1.0 mm) with stereoscopic capability;
0.28 mrad instantaneous field of view (IFOV); 16.8
  16.8

field of view (FOV). Stereo baseline separation of 30 cm.
External calibration target on rover deck.
Thermal Emission Spectrometer
(Mini-TES)
Emission spectra (5?29 mm, 10 cm, 1 resolution) with 8 or
20 mrad FOV. Internal and external blackbody calibration targets.
Instrument Deployment Device (IDD)-Mounted In Situ Package
Alpha Particle X-Ray Spectrometer
(APXS)
244Cm alpha particle sources, and X-ray detectors, 3.8 cm FOV.
Mo?ssbauer Spectrometer (MB) 57Fe spectrometer in backscatter mode; Co/Rh source and Si-PIN
diode detectors; field of view approximately 1.5 cm2.
Microscopic Imager (MI) 30 mm/pixel monochromatic imager (1024  1024) with 6 mm
depth of field.
Rock Abrasion Tool (RAT) Tool capable of brushing or abrading 5 mm deep by 4.5 cm wide
surface on rocks.
Magnets
Filter Located at front of rover within Pancam FOV. Weak magnet to
cull suspended particles from atmosphere and examined by
Pancam, MI, APXS, and MB.
Capture Located at front of rover within Pancam FOV next to filter
magnet. Strong magnet to cull suspended particles from
atmosphere. Examined by Pancam, MI, APXS, and MB.
Sweep Located next to Pancam calibration target. Intended to separate
magnetic from nonmagnetic particles. Examined by Pancam.
RAT Four magnets of different strengths built into RAT. Examined by
Pancam and Hazcam when IDD points RAT toward cameras.
Engineering Cameras
Navigation Cameras
(Navcam)
Mast-mounted panchromatic stereoscopic imaging system with
0.77 mrad IFOV; 45
 FOV, and 20 cm stereo baseline
separation. For planning sequences.
Hazard Avoidance Cameras
(Hazcam)
Front and rear-looking panchromatic stereoscopic imaging
systems with 2 mrad IFOV; 123
 FOV, 10 cm stereo baseline
separation. For path planning and hazard avoidance during traverses.
E12S33 ARVIDSON ET AL.: MER SPIRIT OVERVIEW AND SELECTED RESULTS
7 of 35
E12S33
Table 2. Major Activities for Spirit Organized by Sola
Earth Date Sols Description of Activities Site at Start of Sol
13?14 Jun. 2005 513?514 Drive and ??Methuselah?? Outcrop targeted RS 110
15?27 Jun. 2005 515?527 Drive toward summit of ??Husband Hill??; RS 111?112
28 Jun. to 5 Jul. 2005 528?535 ??Independence?? Outcrop: approach, IDD:
??Franklin,?? ??Jefferson,?? and other targets; RS
112
6?13 Jul. 2005 536?543 Independence Outcrop: scuff, IDD, and
RS ??Penn?? target; ??Independence Panorama??
112
14?20 Jul. 2005 544?549 Drive toward summit of Husband Hill; RS 112
21?27 Jul. 2005 550?556 ??Voltaire??: ??Descartes?? Outcrop:
approach, IDD, and RS ??Discourse?? target
113
27 Jul. to 2 Aug. 2005 556?562 Voltaire: ??Bourgeoisie?? Outcrop: bump,
IDD, and RS ??Chic,?? ??Gentil_Matrice,??
and other targets
113
2?4 Aug. 2005 562?564 Voltaire: Haussmann Outcrop: bump,
IDD, and RS ??Rue_Legendre,??
??Rue_Laplace,?? and ??Sophie_Germain??
113
5?12 Aug. 2005 565?572 Voltaire: ??Assemblee?? Outcrop: bBump,
IDD, and RS ??Gruyere?? targets
113
13?23 Aug. 2005 573?582 Drive toward summit of Husband Hill; RS 113
23?25 Aug. 2005 582?584 Husband Hill Summit Panorama 114
25?29 Aug. 2005 584?588 ??Lambert?? soil: IDD and RS ??Couzy?? and
??Whymper?? targets
114
30?31 Aug. 2005 589?590 Drive toward ??Inner Basin Overlook #1??; RS 114
1?2 Sep. 2005 591?592 Drive toward ??Inner Basin Overlook #2??; RS 114
3 Sep. 2005 to 7 Sep. 2007 593?597 Magnet science; RS 114
8?12 Sep. 2005 598?602 Approach, IDD, and RS ??Irvine?? float target 114
13?21 Sep. 2005 603?611 ??Cliffhanger?? ripple: approach, scuff,
IDD, and RS ??Landsend?? and ??Hang2??
targets, photometry campaign
114
22 Sep. to 4 Oct. 2005 612?623 Drive toward true summit of Husband
Hill; RS; ??Everest Panorama??
114
5?15 Oct. 2005 624?634 ??Hillary?? Outcrop: approach, IDD, and
RS ??Khumjung?? and ??NamcheBazaar?? targets
114
16?22 Oct. 2005 635?641 Drive toward ??Haskin Upper Ridge??; RS 114?118
23?29 Oct. 2005 642?648 ??Kansas?? Outcrop: approach, IDD, and
RS ??Kestral?? target
118
25 Oct. 2005 to 26 Oct. 2006 644?645 Anomaly precludes science activities ?
30 Oct. to 1 Nov. 2005 649?651 Drive toward Haskin Upper Ridge; RS 118
2?5 Nov. 2005 652?654 Remote sensing 118
6 Nov. 2005 655 Drive onto ??Haskin Lower Ridge?? 118
7?8 Nov. 2005 656?657 Remote sensing 119
9 Nov. 2005 658 Drive south toward ??Waypoint??; RS 119
10?14 Nov. 2005 659?663 ??Larry?s Bench?? Outcrop: approach, IDD,
and RS ??Thrasher?? target
119
15?20 Nov. 2005 664?669 Drive south toward ??Waypoint??; RS 119
21?28 Nov. 2005 670?677 ??Seminole?? Outcrop: approach, IDD, and
RS ??Osceola?? and ??Abiaka?? targets;
??Seminole Panorama??
119
29 Nov. to7 Dec. 2005 678?686 Drive toward ??Algonquin?? Outcrop; RS 119
8?10 Dec. 2005 687?689 Algonquin Outcrop IDD and RS ??Iroquet?? target 119
12?18 Dec. 2005 690?696 Drive toward ??Comanche Spur?? Outcrop; RS 119?120
19?25 Dec. 2005 697?703 Comanche Spur Outcrop: approach, IDD,
and RS ??Horseback,?? and ??Palomino?? targets
120
26?27 Dec. 2005 704?705 Drive toward ??El Dorado?? ripple; RS 120
28 Dec. to 1 Jan. 2006 706?710 El Dorado ripple: approach, scuff, IDD,
and RS ??Gallant Knight,?? ??Edgar,?? and
??Shadow?? targets, photometry campaign
120?121
2?10 Jan. 2006 711?719 Drive toward ??Home Plate??; RS 121?122
11?13 Jan. 2006 720?722 Remote sensing 122
14?16 Jan. 2006 723?725 ??Arad?? disturbed soil: IDD and RS
??Samra?? and other targets
122
17?27 Jan. 2006 726?735 Drive toward Home Plate; RS 122?123
25?27 Jan. 2006 733?735 Dynamic brake anomaly and diagnostics ?
28?29 Jan. 2006 736?737 ??BuZhou?? and ??Pan_Gu?? float targets:
IDD and RS
123
30 Jan. to 5 Feb. 2006 738?744 Drive toward Home Plate; RS 123
6 Feb. 2006 745 Remote sensing 124
7?11 Feb. 2006 746?750 ??Barnhill?? Outcrop: approach, IDD, and
RS ??Ace,?? ??Pitcher,?? and ??Fastball?? targets
124
12?15 Feb. 2006 751?754 ??Posey?? Outcrop: approach, IDD, and RS
??Manager?? target
124
16 Feb. 2006 755 Drive toward Home Plate; RS 124
17?19 Feb. 2006 756?758 Remote sensing 124
E12S33 ARVIDSON ET AL.: MER SPIRIT OVERVIEW AND SELECTED RESULTS
8 of 35
E12S33
Table 2. (continued)
Earth Date Sols Description of Activities Site at Start of Sol
20?25 Feb. 2006 759?763 ??Cool_Papa_Bell?? Outcrop: approach,
IDD, RS ??Stars,?? and ??Crawfords?? targets
124
26 Feb. to 1 Mar. 2006 764?767 Drive around Home Plate Rim; RS 124?125
2?5 Mar. 2006 768?771 ??Fuzzy Smith?? float: approach, IDD, and RS 125
6 Mar. to 5 Apr. 2006 772?801 Drive toward ??McCool Hill??; RS 125?126
13?15 Mar. 2006 779?781 Right front drive actuator fault and diagnostics ?
21 Mar. 2006 787 Right front wheel declared nonoperational ?
18 Mar., 24? 25 Mar.,
and 1 Apr. 2006
784, 790?791, 798 RS ??Tyrone?? disturbed soil 126
6?9 Apr. 2006 802?805 Drive toward and approach ??Low Ridge??
location for winter campaign
experiments; RS
126
10?12 Apr. 2006 806?808 Remote sensing 126?127
13?15 Apr. 2006 809?811 ??Enderbyland?? IDD: ??Halley?? Outcrop target 127
16?22 Apr. 2006 812?818 Enderbyland IDD: ??Mawson?? soil target;
Begin McMurdo Panorama
127
22 Apr. 2006 818 RS Tyrone disturbed soil 127
23?25 Apr. 2006 819?821 Remote sensing 127
26 Apr. to 3 May 2006 822?829 Enderbyland IDD and RS ??Progress?? soil target 127
30 Apr. 2006 826 RS ??Tyrone?? disturbed soil 127
4?5 May 2006 830?831 Enderbyland IDD and RS ??Progress1?? target 127
6?13 May 2006 832?838 Enderbyland IDD: ??Halley_Offset?? Outcrop target 127
14?16 May 2006 839?841 Enderbyland IDD: Progress1 brushed soil target 127
17 May 2006 842 IDD positioning test; RS 127
18?19 May 2006 843?844 Remote sensing 127
20?29 May 2006 845?854 Enderbyland: IDD and RS ??Progress2??
brushed soil target
127
24?26 May 2006 850?851 Anomaly precludes science activities ?
30 May to 4 Jun. 2006 855?860 Remote Sensing 127?128
4 Jun. 2006 860 Navcam photon transfer calibration
experiment
128
5 Jun. 2006 861 Enderbyland IDD: ??Halley_Brunt??
Outcrop target
128
6?8 Jun. 2006 862?864 Remote sensing 128
8 Jun. 2006 864 RS Tyrone disturbed soil target 128
9 Jun. 2006 865 Rear Hazcam photon transfer calibration
experiment
128
10?14 Jun. 2006 866?870 Enderbyland IDD: ??Progress3??
brushed soil target
128
12 Jun. 2006 868 Front Hazcam photon transfer calibration
experiment
128
15?19 Jun. 2006 871?874 Remote sensing 128
20 Jun. to 4 Jul. 2006 875?889 Enderbyland IDD: Halley_Brunt Outcrop target 128
21 Jun. 2006 876 Microscopic Imager photon transfer
calibration experiment
128
27 Jun. 2006 882 Left Pancam photon transfer calibration experiment 128
29 Jun. 2006 884 Right Pancam photon transfer calibration experiment 128
5 Jul. 2006 890 Mini-TES elevation actuator calibration 128
6 Jul. 2006 891 Remote sensing 128
7 Jul. 2006 892 Flight software uplink 128
8 Jul. 2006 893 RAT calibration; RS 128
9?11 Jul. 2006 894?896 Remote sensing 128
12?19 Jul. 2006 897?904 Enderbyland IDD: ??Halley_Brunt_Offset1??
Outcrop target
128
19 Jul. 2006 904 RAT unjamming activity 128
20?21 Jul. 2006 905?906 Remote sensing 128
22 Jul. 2006 907 Flight software build; RS 128
24 Jul. 2006 908 Enderbyland IDD: ??Palmer?? ripple target 128
25 Jul. 2006 909 Remote sensing 128
26 Jul. 2006 910 RAT cleaning and calibration; RS 128
27?28 Jul. 2006 911?912 Remote sensing 128
29 Jul. 2006 913 Enderbyland IDD: ??Palmer2??
ripple target
128
30 Jul. to 4 Aug. 2006 914?919 Remote sensing; complete McMurdo Panorama 128
5 Aug. 2006 920 Microscopic Imager photon transfer calibration
experiment; RS
128
6?11 Aug. 2006 921?926 Remote Ssensing; begin filling-in McMurdo
Panorama holes
128
7 Aug. 2006 922 RS Tyrone disturbed soil 128
12?14 Aug. 2006 927?929 Enderbyland IDD: ??Halley_Brunt_Offset2?? target 128
15?21 Aug. 2006 930?936 Remote sensing, including photometry campaign 128
16 Aug. 2006 931 Complete filling-in McMurdo Panorama holes 128
19 Aug. 2006 934 Begin McMurdo Deck Panorama 128
E12S33 ARVIDSON ET AL.: MER SPIRIT OVERVIEW AND SELECTED RESULTS
9 of 35
E12S33
Table 2. (continued)
Earth Date Sols Description of Activities Site at Start of Sol
22 Aug. 2006 937 Enderbyland IDD: Palmer target 128
23/06?26 Aug. 2006 938?941 Remote sensing 128
27 Aug. to 3 Sep. 2006 942?948 Enderbyland IDD: ??Halley_Brunt_Offset3?? target 128
31 Aug. to 1 Sep. 2006 945?946 Anomaly precludes science observations ?
4?9 Sep. 2006 949?954 Remote sensing 128
10?24 Sep. 2006 955?969 IDD capture and filter magnets; RS 128
14 Sep. 2006 959 RS Tyrone disturbed soil 128
17 Sep. 2006 962 Finish McMurdo Deck Panorama 128
20?22 Sep. 2006 965?967 Flight software boot and diagnostics ?
25 Sep. to 31 Oct. 2006 970?1005 MB filter magnet study; RS 128
25 Sep. and 8 Oct. 2006 970, 982 RS Tyrone disturbed soil 128
17?131 Oct. 2006 991?1005 Solar conjunction ?
18, 23, and 31 Oct. 2006 992, 997, 1005 RS Tyrone disturbed soil 128
1 Nov. 2006 1006 IDD capture and filter magnets;
RS Tyrone disturbed soil
128
2?3 Nov. 2006 1007?1008 IDD and RS rock clast targets; IDD Mawson target 128
4?7 Nov. 2006 1009?1012 Remote sensing 128
5 Nov. 2006 1010 Bump bright soil tracks; RS 128
8?11 Nov. 2006 1013?1016 IDD: ??Berkner_Island_1?? disturbed soil target 128
12?17 Nov. 2006 1017?1021 IDD: ??Bear_Island?? disturbed soil target 128
17 Nov. 2006 1021 Pancam calibration target photometry experiment 128
18 Nov. 2006 1022 Bump layered outcrop; RS 128
19?22 Nov. 2006 1023?1026 Remote sensing 128
23?31 Nov. 2006 1027?1035 ??Graham_Land?? Outcrop: IDD and
RS ??King_George_Island?? target
128
30 Nov. 2006 1034 IDD ??Clarence?? and ??Deception?? targets 128
1 Dec. 2006 1036 RS Tyrone disturbed soil; RS 128
3?18 Dec. 2006 1037?1052 Drive toward ??Esperanza?? float; RS 128
13 Dec. 2006 1047 RS Tyrone disturbed soil 128
20?28 Dec. 2006 1053?1061 IDD and RS ??Palma?? target on Esperanza float 128
28 Dec. 2006 1061 RS Tyrone disturbed soil 128
29 Dec. 2006 1062 Drive to location where solar arrays face sun 128
30 Dec. 2006 to 4 Jan. 2007 1063?1068 Atmospheric RS 128
5?25 Jan. 2007 1069?1089 ??Troll?? Outcrop: approach, IDD and
RS ??Montalva?? target
128
16?28 Jan. 2007 1080?1091 Troll Outcrop: RS and IDD ??Riquelme?? and
??Zucchelli?? targets
128
21 Jan. 2007 1085 RS ??Contact?? and ??Londonderry?? targets 128
22 Jan. 2007 1086 IDD ??Svea?? and ??Maudhem?? targets 128
29 Jan. to 9 Feb. 2007 1092?1103 Drive toward Tyrone; IDD and RS
??Mount Darwin?? target
128
9 Feb. 2007 1103 Drive toward Troll 128
10 Feb. 2007 1104?1106 Automode 128
13?19 Feb. 2007 1107?1113 Approach and RS ??Bellingshausen,?? and
??Fabien?? and other outcrops
128
20 Feb. to 10 Mar. 2007 1114?1131 Drive toward Home Plate; IDD and RS;
atmospheric RS
128
11?26 Mar. 2007 1132?1147 Approach, IDD, and RS ??Mitcheltree Ridge??
Outcrop: ??Torquas?? target
128
16 Mar. 2007 1137 MRO safe mode, run-out science submaster executed,
photometry measurements
128
27?28 Mar. 2007 1148?1149 Drive toward Home Plate 128
29 Mar. to 20 Apr. 2007 1150?1171 Drive toward ??Madeline English?? Outcrop;
IDD and RS
128?129
1?8 Apr. 2007 1153?1160 IDD and RS ??Elizabeth Mahon?? target 128
21?30 Apr. 2007 1172?1181 RS and IDD ??Everett,?? ??Slide,?? and ??Good Question??
Outcrop targets
129
1?2 May 2007 1182?1183 RS ??Gertrude_Weise?? disturbed soil 129
3?4 May 2007 1184?1185 Drive toward, IDD, and RS Gertrude Weise 129
5?24 May 2007 1186?1204 IDD ??Kenosha Comets,?? and ??Lefty Ganote?? targets 129
24 May 2007 1204 RS El Dorado ripples 129
25 May to 3 Jun. 2007 1205?1214 RS and IDD??Home Plate Outcrop?? ??Pesapallo,??
??Superpesis,?? and ??June_Emerson?? targets
129
4?9 Jun. 2007 1215?1220 Approach, RS, and IDD Home Plate Outcrop
??Elizabeth Emery,?? ??Jane Stoll,?? ??Mildred Deegan,??
and ??Betty Wagner?s Daughter?? targets
129
10 Jun. 2006 to 23 Jun. 2007 1221?1234 Approach, IDD, and RS ??Nancy Warren??
Outcrop and target
129?130
13 Jun., 22 Jun., and 12 Jul. 2007 1224, 1233, 1252 Dust cleaning events 130
23 Jun. to 6 Jul. 2007 1234?1246 IDD and RS ??Eileen Dean?? disturbed soil;
atmospheric RS
130
30 Jun. to 3 Jul. 2007 1240?1243 Stand down; atmospheric RS 130
E12S33 ARVIDSON ET AL.: MER SPIRIT OVERVIEW AND SELECTED RESULTS
10 of 35
E12S33
for changes relative to measurements done before arriving at
Low Ridge for the winter. After finishing these measure-
ments, the rover focused on remote sensing and in situ work
on the finely layered outcrops on the western side of
Mitcheltree Ridge, most notably Torquas (rock class defined
by high K2O, Zn, Ni contents and enrichment in magnetite
[Morris et al., 2008; Ming et al., 2008]). Spirit was then
commanded to approach and make remote sensing and in
situ measurements on outcrops just to the east of Home
Plate, including Elizabeth Mahon (new class of rock char-
acterized by high SiO2 content [Squyres et al., 2008; Ming
et al., 2008]), Madeline English, Everett and Good Question
(latter two rocks define new classes on the basis of high
MgO and magnetite contents for the former, and high SiO2
and low MnO contents for the latter [Morris et al., 2008;
Ming et al., 2008]). During one of its backward drives Spirit?s
right front wheel excavated light-toned soil deposits (Ger-
trude Weise), which Mini-TES indicated and APXS meas-
urements later confirmed have extraordinarily high silica
contents [Squyres et al., 2008]. After completing measure-
ments on these unusual deposits the rover drove to acquire
detailed remote sensing and in situ measurements of the
strata exposed on the eastern flank of Home Plate on sols
1205 to 1220 (Pesapallo, Superpesis, June Emerson, Eliz-
abeth Emery). An outcrop identified as silica-rich from
Mini-TES observations was then examined (Nancy War-
ren), followed by another bright deposit excavated by the
right front wheel (Eileen Dean). The vehicle was then
commanded to drive over and crush rocks adjacent to
Nancy Warren and broke open and obtained in situ obser-
vations on the silica-rich targets Innocent Bystander and
Norma Luker.
[10] During an approximately 60 sol period when the
vehicle was parked over Innocent Bystander waiting out the
low solar energy conditions associated with a southern
hemisphere dust storm, one or more of the Mini-TES
mirrors was significantly contaminated by dust, reducing
the spectral sensitivity of this instrument. After the dust
storm, Spirit was commanded to approach and ascend onto
Home Plate and obtain measurements designed to charac-
terize the rocks on the top of Home Plate and rocks on the
South Promontory. With the third winter season approach-
ing, an increasing cover of dust on the solar panels, and
hindered mobility on slopes due to the inoperative right
front wheel, a decision was made to drive to the northern
flank of Home Plate to achieve a maximum possible
Earth Date Sols Description of Activities Site at Start of Sol
7 Jul. to 21 Aug. 2007 1247?1291 Nancy Warren Outcrop IDD ??Innocent Bystander??
and ??Norma Luker ??targets; RS,
including photometry campaign
130
9?20 Aug. 2007 1279?1290 MI diagnostics 130
22 Aug. to 3 Sep. 2007 1292?1304 RS and drive toward Home Plate
Mini-TES recalibration using ??Gertrude Weise??
disturbed soil
130
4?7 Sep. 2007 1305?1308 Drive toward Home Plate 130
7 Sep. 2007 1308 Reach Home Plate, RS 130
9?10 Sep. 2007 1309?1310 RS Home Plate 130
10?14 Sep. 2007 1310?1314 Drive toward Home Plate Site 2; RS 130?131
15 Sep. to 7 Oct. 2007 1315?1337 Drive toward Home Plate Site 3; RS;
??Home Plate South Panorama??
131
25?30 Sep. 2007 1325?1330 IDD ??Texas Chili?? Outcrop target 131
16 Oct. 2007 1337?1345 Home Plate Site 3a feature: IDD
??Humboldt Peak?? float target
131
17?23 Oct. 2007 1346?1352 Drive toward Home Plate Site 4; RS 131
17 and 26?31 Oct. 2007 1346, 1355?1360 RAT diagnostics; RS 131?132
19?23 Oct. 2007 1348?1352 Long-baseline stereo off ??Home Plate?? 131
24?25 Oct. 2007 1353?1354 Drive toward Home Plate south off-ramp; RS 131
1?3 Nov. 2007 1361?1363 Drive toward Home Plate Site 5; RS 132
4?10 Nov. 2007 1364?1370 IDD ??Pecan Pie?? Outcrop target; RS 132
11?14 Nov. 2007 1371?1374 Drive toward Home Plate Site 6 132
15?20 Nov. 2007 1375?1380 Drive toward Home Plate Site 7; RS 132
20 Nov. to 19 Dec. 2007 1380?1408 Drive toward winter campaign site;
atmospheric RS
132?133
20 Dec. to 15 Jan. 2008 1409?1434 IDD ??Chanute?? Outcrop target;
Tuskegee Panorama
133
16?20 Jan. 2008 1435?1439 RS Fuzzy Smith float; atmospheric RS;
RAT imaging
133
21 Jan. to 23 Mar. 2008 1440?1500 Atmospheric RS 133?134
26 Jan. to 6 Feb. 2008 1445?1456 IDD ??Freeman?? Outcrop target; RS
??Fuzzy Smith 2 Float,?? ??Winston Gaskins 2??
Outcrop target
133
26 Feb. 2006 to 20 Mar. 2008 1475?1497 IDD and RS ??Wendell Pruitt?? Outcrop
and ??Arthur C. Harmon?? Soil targets;
RS ??C.S. Lewis?? and Freeman Outcrop targets
134
27 Feb. 2008 1476 Begin Bonestell Panorama 134
aMeasurements were made on rock outcrops, float rocks, soils, and soils disturbed by the vehicle?s wheels. RS, remote sensing; IDD, Instrument
Deployment Device. Note that the Mini-TES instrument capabilities were affected by dust on the mirror on sol 420, followed by additional accumulation
during the course of the extended mission. Other acronyms defined in Table 1.
Table 2. (continued)
E12S33 ARVIDSON ET AL.: MER SPIRIT OVERVIEW AND SELECTED RESULTS
11 of 35
E12S33
northerly tilt to survive the winter season. Spirit reached the
site on sol 1408 and began a set of maneuvers to bump
down the flank to follow the sun as it moved north toward
its solstice position. Remote sensing and in situ measure-
ments continued to be made as the vehicle moved down-
slope, contingent upon available power.
3. Husband Hill: Voltaire Outcrop Campaign
[11] The first scientific highlight to be discussed is the
discovery of the layered Voltaire outcrops encountered
after Spirit left the Independence Outcrop and continued
its ascent to the summit of Husband Hill (Figure 2b and
Table 3) [also see Clark et al., 2007]. Outcrops observed
prior to discovery of the Voltaire sequence showed that
bedding was largely conformable with the topography of the
Hills, including outcrops on West Spur and the ensemble of
outcrops found in Larry?s Lookout on the northwest flank of
Husband Hill (Appendix A) [Squyres et al., 2006]. The
observation that the outcrop pattern on Husband Hill con-
forms to topography implies that the Hill is the geomorphic
expression of an antiformal structure and that Spirit has
largely examined the top of the section [McCoy et al.,
2008]. On the other hand, the end-of-drive Pancam mosaic
acquired on sol 549 showed a set of strata approximately
5 m long in a NE?SW direction and 2.5 m wide in a NW-
SE direction, with a dip to the SE, i.e., dipping into the hill
(Figure 4). The orientation of the strata was verified using a
CAVE Virtual Reality Immersion System to view the out-
crops at full scale and in three dimensions using Pancam
and Navcam mosaics. This outcrop, which was named
Voltaire because it was discovered on Bastille Day (14 July
2005), is truncated to the SW by a shallow trough and to the
Table 3. Summary of the Voltaire Outcrop Measurement Campaign
Sols Description of Activities
552 Started IDD work on Discourse target located on Descartes Outcrop with a
MI mosaic and APXS integration.
553 Brush, MI, and APXS on Discourse.
554?555 MB integration on Discourse.
556 MI on two additional Descartes targets: Petitchou and Moncherie
(dark clast). At end of sol the rover turned to face the rocks Sourir,
Bourgeoisie, and Haussmann.
557 Started IDD work on Bourgeoisie Outcrop, with three MI mosaics and an
APXS integration. Targets Gallant, Gentil, and Chic (APXS target), clast
embedded in the outcrop.
558?559 MB integration on Chic target (Bourgeoisie)
560 Acquired MI, RAT brush, MI, and APXS data on Gentil Matrice target
(Bourgeoisie).
561 MB integration on Gentil Matrice target.
562 Short MB integration on Gentil Matrice target. Sol finished with the rover
turning to face Haussmann.
563 Three MI mosaics acquired on Haussmann Outcrop for targets Rue Legendre,
Rue Sophie Germain, and Rue Laplace. An APXS measurement
acquired on Rue Laplace.
564 Sol activities were lost when uplink did not make it to the rover.
565 Short drive to Assemblee.
566 MI and APXS integration on Gruyere target (Assemblee Outcrop).
Target was too rough for RAT brush.
567?570 Four sols of MB integration on Gruyere target because of low Fe content.
571 MI and APXS integration on second Gruyere target to test
whether clasts were source of high Cr in initial APXS measurement.
Target name Gruyere_APXS.
572 IDD stowed in preparation of driving.
573 Drove away from Voltaire outcrops, looking back with
Pancam during middrive.
Figure 4. Portion of the sol 549 Pancam drive direction
mosaic showing layered strata forming the Voltaire out-
crops. The rocks strike from lower right to upper left
(northeast to southwest) and dip into the hill (toward the
southeast). Bourgeoisie, Haussmann, Descartes, and As-
semblee are outcrops that were the focus of detailed in situ
measurements by Spirit. Cocarde appears to have broken
from Assemblee and moved slightly down hill. Both of
these latter rocks have similar Mini-TES spectral emissivity
signatures indicative of glassy materials. Enlightenment,
Femey, and Revolution have emissivity signatures similar to
one another and the rest of the Voltaire outcrops and also
appear to be dominated spectrally by glassy materials.
Pancam mosaic 2PP549ILFADCYL00P2352L777M2.
E12S33 ARVIDSON ET AL.: MER SPIRIT OVERVIEW AND SELECTED RESULTS
12 of 35
E12S33
NE by a cover of soil and rocks (Figure 4). Specifically,
detailed image analyses demonstrated the presence of a half
dozen discrete layers striking 
60
 clockwise from north
and dipping to the southeast (Table 4). The Voltaire out-
crops may be a discrete me?lange block emplaced by impact
or tectonic movements, or a tilted section of a more
extensive exposure of bedrock.
[12] An extensive campaign was undertaken by Spirit in
which a number of targets on outcrops (Descartes, Bour-
geoisie, Haussmann, and Assemblee) were examined using
both the remote sensing and in situ instruments (Table 3).
The outcrops exhibit two basic morphologic patterns. Des-
cartes, Bourgeoisie, and Haussmann are characterized by
relatively smooth, tabular appearances with well-defined
bedding and joint planes (Figure 5). A number of sub-
rounded to rounded clasts ranging in size from a centimeter
to a few centimeters are contained in these tabular outcrops,
i.e., the deposits are conglomerates (Figure 5). The exposed
clast surfaces have been shaped into ventifacts by wind
erosion. The Chic clast on Bourgeoisie was the target of
Microscopic Imager (MI), APXS, and MB observations,
along with nearby matrix materials (Discourse and Gentil
Matrice) (Tables 1, 3, 5, and 6). Additional matrix measure-
ments were made on the Haussmann and Descartes out-
crops. The second morphologic form is represented by the
top-most portion of the Voltaire Outcrop (Assemblee;
Tables 3?5 and Figure 6). This conglomeratic rock exhibits
a nodular or crumbly appearance and when examined with
the MI shows numerous relatively small and well rounded
clasts (typically 
< 0.5 cm across) (Figure 6b) [Clark et al.,
2007]. Several other rocks with nodular textures similar to
those for Assemblee can also be seen (e.g., Cocarde,
Egalite; Figure 4), although as float rather than as outcrop.
Assemblee was the only rock of this type for which MB,
APXS, and MI data were acquired.
[13] Pancam multispectral observations were obtained for
the Voltaire outcrops and surrounding rocks and soils and
representative spectra are shown for Descartes and Assem-
blee surfaces in Figure 7. The absorption and scattering
features for the spectral range covered by Pancam (0.4 to
1.0 mm) are controlled to first order by the abundance and
textural characteristics of ferrous and ferric bearing minerals
[e.g., Burns, 1993], including pyroxene, olivine, nanophase
iron oxides, hematite, and goethite, all of which have been
detected by Spirit?s MB during its measurements on the
Columbia Hills [Morris et al., 2006, 2008]. The Pancam
spectrum for the Descartes surface is brighter than the one
for Assemblee and shows a steeper ferric absorption edge
Table 4. Summary of Strikes and Dips Calculated for Voltaire Outcropsa
Sol and Frame Number Strike (degrees) Dip to Southeast (degrees)
565 2R176524254RSDADAEP1312L0MZ N45E 35
565 2R176523021RSDA2P1301L0MZ N29E 23
565 2R176524254RSDADAEP1312L0MZ N48E 37
563 2F17634947RSLAD92P1121LOMZ N75E 46
aOrientations for the bedding planes were derived from unit vector surface normal data computed from stereo observations
using regions of interest defined by inspection to isolate bedding surfaces.
Figure 5. Pancam false color composite of the Descartes Outcrop oriented so that the bedding planes
are parallel to the bottom of the image. Joints cut across the outcrop from lower right to upper left and can
also be seen in the sol 549 approach mosaic (Figure 4). Locations of in situ observations are shown,
including the Moncherie clast (bottom most labeled clast). Pancam frames 2P175278077ES-
FAD40P2558L2M1, 2P175278161ESFAD40P2558L5M1, and 2P175278228ESFAD40P2558L7M1
were used to generate the mosaic.
E12S33 ARVIDSON ET AL.: MER SPIRIT OVERVIEW AND SELECTED RESULTS
13 of 35
E12S33
between 0.4 to 0.8 mm. This is consistent with the MB and
APXS observations that together show a threefold higher
concentration of nanophase iron oxides and the presence of
goethite (FeOOH) in Descartes as compared to Assemblee
(Tables 5?6). The Descartes spectra show a band minimum
between 0.9 to 1 mm, whereas the spectra for the Assemblee
surface are relatively flat in this wavelength region. These
patterns are consistent with the higher proportion of Fe-
bearing minerals in Descartes (14.2% versus 6.4%, as FeO)
based on APXS observations (summarized in Table 6)
[Ming et al., 2008] and the detection of iron-bearing
pyroxene for both outcrops from MB data (Table 5).
Although great care must be exercised in comparing Pan-
cam spectra with MB and APXS data because of the
different sampling depths in this case the patterns in the
data imply that all three instruments are sampling the same
materials.
[14] Mini-TES observations were acquired for a variety of
rock targets during the extensive campaign of in situ
observations on the Voltaire outcrops (Figures 8 and 9).
The Mini-TES observations illustrate the synergistic use of
the remote sensing elements of the Athena Payload (Table
1) in that the targeting for the observations took advantage
of both Navcam and Pancam (including color) mosaics to
cover the range of morphologic forms for the outcrops and
float rocks within Spirit?s field of view. Figure 8 shows the
Mini-TES ??footprints?? projected onto a portion of a Nav-
cam mosaic taken when Spirit was sitting on the Voltaire
outcrops. On the basis of emissivity patterns the rocks can
be divided into five distinct spectral classes (Figure 9). First,
many of the boulders uphill from the Voltaire Outcrop are
spectrally similar to the Wishstone float rock observed
earlier in the mission and found in many places on Husband
Hill, as determined from Pancam and Mini-TES observa-
tions (see Appendix A for Wishstone float rock location)
[Farrand et al., 2006; Ruff et al., 2006]. The Wishstone-like
emission spectra are dominated by stretching and bending
vibrations of plagioclase feldspar of an intermediate com-
position, as shown in Figure 9. A second and relatively rare
class consists of boulders with spectra similar to the basaltic
rock, Backstay (encountered earlier on Husband Hill, see
Appendix A for location), based on olivine and pyroxene
stretching and bending vibrational modes (Figure 9). A third
and also relatively rare class of float rock is spectrally
similar to the olivine-bearing basalt rocks that dominate
the plains (i.e., similar to Adirondack and represented by
Liberte; Figures 8 and 9). The abundance of Wishstone
materials and their altered equivalents, Watchtower materi-
als [e.g., Ming et al., 2008], on Husband Hill implies that
the top of the antiformal stratigraphic section is dominated
by these materials. Mapping by us shows that the Voltaire
Outcrop is stratigraphically below Wishstone materials and
includes the nearby Independence Outcrop (Appendix A).
[15] Emissivity spectra for the Voltaire outcrops show a
fourth and fifth pair of spectral classes that are distinctly
different from those discussed in the previous paragraph. In
particular, the Descartes class is characterized by broad,
relatively featureless absorption in the low-wave-number
region (<600 cm1) and dominates the spectral appearance
for Descartes, Bourgeoisie, and Haussmann Outcrop obser-
vations. The spectra are consistent with the presence of an
amorphous silicate phase and also share similarities with the
typical ??dust?? spectra identified in many places using Mini-
TES data. Spectra for Assemblee form the fifth class and
show a distinct emissivity minima at 
1050 cm1 and

460 cm1 (Figure 9) and are similar to spectra for the
outcrop, Clovis, on West Spur (see Appendix A for Clovis
location and Ruff et al. [2006] for further spectral details).
The broad nature of these two emissivity minima is similar
to the spectra of glassy or amorphous silicate phases [e.g.,
Parke, 1974]. There are also spectra for float rocks near the
Voltaire outcrops that match a mix of materials and are
thought to represent Voltaire Outcrop materials that have
been weathered and left exposed as boulders (Figure 8).
[16] The Mini-TES footprint is too large to isolate Vol-
taire Outcrop clast material. However, MB and APXS
observations provide compelling evidence that the clasts
are dominated by Wishstone-type materials (Tables 5 and 6).
Table 5. MB-Based Percentage Iron Mineralogy for Key Targets From Morris et al. [2006, 2008]
Name Olivine Pyroxene Ilmenite Chromite
Nanophase
Iron Oxide Magnetite Hematite Goethite Fe+3/Fetotal
Descartes (A555RB0) 1 27 0 0 43 18 5 7 0.68
Chic (A559RB0) 26 25 18 0 23 7 2 0 0.31
Assemblee (A568RU0) 0 44 0 23 32 0 0 0 0.37
Cliffhanger (A609SU0) Cliffhanger_Hang2 13 41 0 0 30 7 9 0 0.45
El Dorado (A708SU0) Shadow 47 32 0 0 8 12 0 0 0.17
Halley (A836RU0) Halley_Offset 2 3 0 0 8 15 73 0 0.88
Esperanza (A1056RU0) Palma 4 45 0 0 4 45 1 0 0.40
Table 6. APXS Compositional Data From Ming et al. [2006, 2008]
Oxides (wt %) Elements (mg/kg)
SiO2 TiO2 Al2O3 FeO MnO MgO CaO Na2O K2O P2O5 Cr2O3 Cl SO3 Ni Zn Br Ge
A552_OU_Descartes_Discourse 45.32 0.97 9.93 14.29 0.24 9.35 5.56 3.14 0.61 1.38 0.17 1.30 7.64 436 209 155 15
A566_OU_Assemblee_Gruyere 51.01 0.79 17.39 6.43 0.15 8.21 3.77 1.70 0.93 1.60 2.86 0.92 4.03 1248 244 65 22
A611_SU_Cliffhanger_Hang Two 47.73 1.20 12.34 10.77 0.22 7.24 7.13 3.60 0.51 2.10 0.13 0.78 6.16 168 155 104 4
A709_SU_El Dorado_Shadow 46.91 0.62 10.74 15.96 0.31 11.31 6.10 3.01 0.31 0.81 0.32 0.38 3.06 997 114 22 0
A833_SU_Enderbyland_Halley Center 45.30 0.90 8.73 17.97 0.24 9.31 5.30 2.71 0.60 0.90 0.19 0.86 6.67 777 2270 32 17
A1055_RU_Esperanza_Palma 47.9 1.05 8.40 20.2 0.38 8.45 5.57 3.40 0.52 0.91 0.20 0.47 2.36 395 368 181 19
E12S33 ARVIDSON ET AL.: MER SPIRIT OVERVIEW AND SELECTED RESULTS
14 of 35
E12S33
This conclusion is strengthened by use of correspondence
analysis applied to APXS elemental compositional data for
rocks from the hills (Figure 10). Correspondence analysis is
a powerful technique for exploring structure and relation-
ships among samples and variables in a multidimensional
data set. It is a form of principal component analysis in
which normalization of the data matrix allows plotting of
factor loadings for both samples and variables using the
same scales [e.g., Arvidson et al., 2006a, 2006b]. In
particular, analyses for the matrix material on Descartes,
Bourgeoisie, and Haussmann show that these targets plot
close to compositional origin (i.e., the average) of rocks
analyzed in the Hills. The Chic clast (in the Bourgeoisie
Outcrop) has a composition displaced toward the Wishstone
sample location when projected onto the first two factor
vectors (Figure 10). It also shows that Assemblee is
enriched in Cr2O3, SiO2, K2O, and Al2O3 and depleted in
FeO relative to the matrix material in the other Voltaire
outcrops. In fact, removal of Cr2O3 and recomputing the
factor loadings show that Assemblee and Independence
have compositional similarities, a pattern also noted by
Clark et al. [2007]. Assemblee is also enriched in Ni and
Ge relative to the other matrix materials in the Voltaire
outcrops (Table 6).
[17] The amorphous material in the Voltaire outcrops
detected by Mini-TES cannot be iron-bearing glass since
this material would have also been detected by the MB
instrument for Descartes, Bourgeoisie, and Haussmann, and
Assemblee (and Clovis) and was not (Table 5) [Morris et
al., 2006, 2008]. Rather the iron-bearing minerals in the
outcrops (with the exception of clasts) are dominated by
nanophase iron oxides and pyroxene. The Descartes matrix
also has magnetite, hematite, and goethite present whereas
Assemblee has chromite as a third phase.
[18] The ensemble of data collected for the Voltaire
experiments allows development of a working hypothesis
for the formation and alteration of these outcrops. The
conglomeratic nature of the outcrops, combined with the
evidence for an amorphous or glassy component, argues for
emplacement as an impact melt that incorporated local
Wishstone materials, although emplacement as an energetic
volcanic flow or pipe cannot be ruled out. Postemplacement
alteration in an aqueous environment is indicated by the
detection of goethite (FeOOH) in the Descartes Outcrop
(Table 5). The presence of nanophase iron oxides, hematite,
and goethite in Descartes and nanophase iron oxides in
Bourgeoisie, and Haussmann outcrops, combined with the
fact that none of the glassy or amorphous phases detected
by Mini-TES are ferrous- or ferric-bearing, provide addi-
tional clues for the alteration history of these materials. We
envision a process of in situ alteration of the glassy phase in
these rocks, releasing iron to form iron oxides and goethite,
not unlike a terrestrial process in which allophane and iron
oxides and oxyhydroxides are generated in soils weathered
from basaltic rocks [Parfitt and Furkert, 1980] and glassy
basaltic tephra [e.g., Morris et al., 2000]. Chromite in
Assemblee (Table 5) may have remained immobile during
this process. The Si and Ge enhancements and low FeO in
Assemblee relative to the rest of the Voltaire measurements
(Table 6) are consistent with the trends seen in Hawaiian
soils formed from relatively intense aqueous weathering of
basaltic rocks, i.e., SiO2 and Al2O3 and Ge concentrate in
the soils while FeO is removed by aqueous processes
[Morris et al., 2000; Kurtz et al., 2002; Ming et al.,
2008]. Finally, we note that Schmidt et al. [2008] concluded
from analyses of Mini-TES spectra that glassy or amor-
phous components are indeed widespread in the Columbia
Hills sites visited by Spirit, although no iron bearing glassy
phases have been detected by MB [Morris et al., 2008].
Figure 6a. Pancam false color view of the Assemblee
Outcrop showing the target Gruyere that was the focus of
in situ measurements. Note the friable nature of this
outcrop. The rock strikes from lower right to upper left
(SW toNE) and dips into the hill (toward the SE). The Pancam
frames used were 2P175540927ESFAD56P2566L2M1,
2P175541009ESFAD56P2566L5M1, and 2P175541077ES-
FAD56P2566L7M1.
Figure 6b. MI frame of Gruyere showing the presence of
rounded, embedded clasts. Frame covers 
3 cm across.
Frame number 2M176609837EFFADAEP2936M2F1.
E12S33 ARVIDSON ET AL.: MER SPIRIT OVERVIEW AND SELECTED RESULTS
15 of 35
E12S33
Thus the process described above may have been a common
and widespread phenomenon that altered glassy deposits.
4. Inner Basin: Volcanic Rocks and Associated
Deposits
[19] Spirit entered the Inner Basin of the Columbia Hills
after descending Husband Hill and Haskin Ridge, spending
>800 sols exploring the landforms and deposits in and
around the 
80 m wide ovoidal plateau termed Home Plate
(Figures 2a?2d and Table 2). Rock exposures on the flanks
of Home Plate are dominated by volcanic tuff deposits that
show evidence for energetic explosive emplacement, in-
cluding cut and fill structures, cross bedding, and a ??bomb
sag?? produced when a block was ejected into the atmo-
sphere and impacted into deformable tuff deposits [Squyres
et al., 2007; Lewis et al., 2008]. Interaction of subsurface
magma with groundwater is the likely cause of the explo-
sive volcanism.
[20] Mitcheltree Ridge and Low Ridge (Figures 2 and
11), located to the east and southeast of Home Plate,
respectively, are capped by vesicular basalt boulder fields
and wind-blown soils (Figure 11). The rock, Esperanza,
examined by Spirit after leaving the Low Ridge Winter
Campaign site, is an example of one of these boulders
(Table 2 and Figures 11?13). Examination of the MI
coverage of Esperanza shows that the vesicles have been
finely shaped by wind, forming a series of sharp edges
typical of wind-sculpted vesicular basalt outcrops and
boulders on Earth (Figure 12b). In fact the MI data show
sand, the likely abrasive agent, sitting in one of the vesicles.
MB observations of Esperanza indicate subequal amounts
of iron within pyroxene and magnetite, with a minor
component in nanophase iron oxides (Table 5). Pancam
spectra for Esperanza are consistent with the presence and
relative abundances of these minerals in that the spectra are
dark (consistent with relatively high magnetite concentra-
tion) and show a slight downturn at longer wavelengths
consistent with the presence of pyroxene (Figure 7). Mini-
TES spectra are dominated by the presence of pyroxene
bending and stretching mode vibrations [see also Schmidt et
al., 2008] and are similar to the spectra for Bounce Rock, a
pyroxenite examined in detail by Opportunity in Meridiani
Planum (Figure 13).
[21] Thinly bedded, platy outcrops of granular materials
composed of sand-sized grains are found near the bottom of
both Low Ridge and Mitcheltree Ridge and the outcrop,
Troll, located between the two Ridges (Figures 11 and 2d).
Detailed mapping using Navcam panoramic images ac-
quired from the Low Ridge winter campaign site, during
traverses to and from Tyrone, and while Spirit was located
in the Eastern Valley between Home Plate and the Ridges
(Figure 2d), shows that these platy deposits are found on
both western and eastern sides of the ridges and are
dominated by dips into the Ridges [Lewis et al., 2007].
The deposits on the northern side of Low Ridge define dips
Figure 7. Pancam spectra for a series of features discussed in this paper. The legend at the top of the
figure denotes the feature names. Numbers in parentheses represent the Fe+3/Fetotal ratios derived from
MB measurements for targets on the relevant features. The brighter spectra have higher ratios. Error bars
represent 1 standard deviation about the mean values for the group of pixels used to generate the spectra.
All the spectra show a strong ferric oxide absorption edge typical for Mars. Longer-wavelength features
are also diagnostic of mineralogy and are discussed in detail in the text. R* is equivalent to Lambert
albedo. The Pancam frames used are as follows: Descartes (p2558, Sol 551), 2P175278077IO-
FAD40P2558L2C1; Assemblee (p2541, Sol 572), 2P177142010IOFADAEP2541L2C1; El Dorado
(p2536, Sol 711), 2P189482458IOFAL02P2536L2C1; Cliffhanger (p2587, Sol 611), 2P180607674IO-
FAEM9P2587L2C1; Halley (p2286, Sol 925), 2P208478741IOFAS00P2286L2C1; Esperanza (p2599,
Sol 1070), 2P221351648IOFASCGP2599L2C2.
E12S33 ARVIDSON ET AL.: MER SPIRIT OVERVIEW AND SELECTED RESULTS
16 of 35
E12S33
that wrap around and form the northern nose of a synclinal
structure. Thinly bedded, platy outcrops are also evident on
the eastern flank of Home Plate and dip to the west. Thus
the outcrops on the Eastern Valley floor between Home
Plate and the Ridges are exposed as the core of an
antiformal structure.
[22] The dominant materials exposed in the Eastern
Valley, in addition to wind-blown soils, are platy, buff
colored outcrops. Halley is one of these outcrops and was
examined in detail while Spirit was parked for the winter at
Low Ridge (Table 2 and Figure 11). MI images show that
this platy outcrop is fine grained. MB data show that
Halley?s iron-bearing minerals are dominated by hematite
and that this outcrop has a very high Fe+3/Fetotal ratio
(Table 5). Pancam spectra extracted for Halley and sur-
rounding platy outcrops show a high overall albedo, pro-
nounced ferric slope between 0.4 and 0.8 mm, an increasing
reflectance with increasing wavelength out to 
0.9 mm, a
minor inflection at 
0.9 mm, and a relatively sharp negative
spectral slope between 
0.9 and 1.0 mm (Figure 7). The
sharp down turn for light-toned disturbed soils in the
Eastern Valley observed by Pancam has been interpreted
by Rice et al. [2008] as an O-H stretch overtone or a water
combination band resulting from the presence of hydrated
or hydroxlated mineral phases. On the basis of extensive
APXS observations, Halley may contain calcium sulfate
minerals, e.g., gypsum or anhydrite [Yen et al., 2008; Ming
et al., 2008]. The ensemble of evidence for Halley implies
that hermatite and hydrated or hydroxlated sulfate-bearing
phases are present.
[23] The thinly bedded, platy outcrop, Graham Land
(King George Island target), is located stratigraphically
above the Halley deposits and was examined by Spirit as
it left the winter campaign site (Figure 11). MI data show
that this outcrop is dominated by round, very coarse sand-
sized grains, and MB data show that hematite dominates the
iron-bearing mineralogy (Table 5 and Figure 14). Additional
and similar deposits (Montalva and Riquelme) were exam-
ined by Spirit when it parked at the adjacent Troll Outcrop
and the ensemble of data again show that these deposits are
platy, granular, and that iron-bearing materials are domi-
nantly in hematite. In fact, correspondence analysis applied
to MB data for all rocks found within the Inner Basin shows
a progressive enrichment in hematite relative to other iron
bearing minerals for the platy outcrops found on the Ridges
and within the Eastern Valley (Figure 15).
[24] As noted, the platy outcrops exposed in the Eastern
Valley between Home Plate and the Ridges are largely
covered by a thin veneer of wind-blown soil. In some
locations the platy outcrops are covered not by soils but
by clasts that exhibit a nodular appearance (Figure 16).
These nodular materials, where examined by Spirit?s Mini-
TES and APXS instruments, were found to be outcrops of
silica-rich materials or debris fields of the same type of
Figure 8. A subset of the Navcam mosaic acquired on sol 551 looking to the northeast, with color-
coded footprint locations for acquisition of Mini-TES spectra overlain. Green colors correspond to the
Descartes class spectrum shown in Figure 9. Magenta corresponds to the Assemblee spectral class. Red
corresponds to a Wishstone-like, orange corresponds to a spectral mix of Assemble and Wishstone, and
blue corresponds to an Adirondack-like spectrum. The Backstay spectral class is not shown on this
mosaic. Note the dominance of Wishstone-like materials toward the uphill section of the image. Navcam
mosaic 2NN551EFFADCYL56P0660L00M1.
E12S33 ARVIDSON ET AL.: MER SPIRIT OVERVIEW AND SELECTED RESULTS
17 of 35
E12S33
materials crushed by the vehicle?s inoperative right-front
wheel [Squyres et al., 2008, Figure 16]. Mini-TES spectra
indicate the presence, in particular, of opaline silica, and
APXS data for the Gertrude Weise disturbed soils show that
the material is almost pure silica (
90%) with the remaining
component consisting largely of TiO2 [Squyres et al., 2008,
Figure 17]. These materials are inferred to have been
deposited by aqueous fluids under acid-sulfate low pH
conditions by either dissolution and reprecipitation in-place
or transport in solution to the depositional location [Squyres
et al., 2008]. Additional support for the presence of acid-
sulfate fluids is the presence of hydrated sulfate deposits in
soils disturbed by Spirit?s wheels in the Inner Basin to the
north of Home Plate (Arad feature) and to the east of
Mitcheltree Ridge (Tyrone feature; Figures 2 and 11).
Emissivity spectra indicate the presence of hydrated materi-
als based on the presence of a 
6 mm feature attributed to
the fundamental bending mode of water, MB data show the
presence of a ferric-sulfate phase, and APXS data show
that SO3 increases with increasing SiO2 concentrations
(Figure 17) [Morris et al., 2008; Yen et al., 2008; Wang et
al., 2008].
[25] The results from the data collected by Spirit in and
around Home Plate allow development of a model that
focuses on the Inner Basin volcanic history and the role of
aqueous fluids in emplacement and modification of the
deposits. As noted, Home Plate has been interpreted to be
a partially eroded volcaniclastic construct [Squyres et al.,
2007]. Acid sulfate aqueous activity (vapor and/or fluids)
occurred as the volcaniclastic deposits formed, interleaving
silica-rich and sulfate-rich deposits, and altering some of the
ash materials to hematite and other phases. For most of its
history wind erosion has probably dominated changes in the
Inner Basin, differentially stripping the materials to produce
the landforms, outcrops, and soil exposures explored by
Spirit.
[26] The Kau Desert, Kilauea, Hawaii is offered as an
Earth analog for the generation of volcaniclastic materials,
lava flows, and formation of silica-rich and sulfate-rich
deposits in the Inner Basin (Figure 18). In the Kau Desert,
basaltic ash deposits are found interleaved with basaltic lava
flows. Some of the ash deposits formed as rounded, well
sorted, and oxidized accretionary lapilli produced when wet
ash clouds caused nucleation of fine airborne ash into
spherical sand grains. This is a plausible explanation for
the finely layered, platy, granular deposits evident in the
King George Island Outcrop and elsewhere in the Inner
Basin ridges (Figure 14). Opaline silica encrustations are
common in the Kau Desert and form in the vicinity of steam
vents by dissolution of rock and in-place precipitation of
opaline silica, and as case-hardened, but friable coatings on
ash deposits created as silica-rich solutions evaporated when
they reached the surface [Malin et al., 1983; Schiffman et
al., 2006; K. Seelos et al., Silica in a Mars Analog
Environment: Ka?u Desert, Kilauea Volcano, Hawaii, man-
uscript in preparation, 2008]. Further, in some locations
acid-sulfate rich solutions have thoroughly altered basaltic
materials and left behind silica-rich residues. Sulfate depos-
its (e.g., primarily gypsum, minor amount of jarosite) also
form near steam vents, although the minerals are retained
primarily where protected (beneath overhangs or within
internal cavities within lava flows) from rainfall and disso-
lution (Seelos et al., manuscript in preparation, 2008).
5. Coordinated Analyses of Orbital and Spirit-
Based Data Sets
[27] The previous sections of this paper have highlighted
the extensive exploration and measurement campaigns
conducted by Spirit in the Columbia Hills and focused on
Figure 9. Mini-TES spectra (in black) from the Voltaire
region compared with laboratory and other Mini-TES
spectra in colors. Descartes spectra have a significant dust
component as evident from the scaled Gusev dust spectrum
(brown) and the relatively featureless low-wave-number
region (<600 cm1) is consistent with a mixture containing
an amorphous silicate phase like basaltic glass (magenta;
scaled laboratory). Assemblee spectra are dominated by an
amorphous silicate phase like basaltic glass (magenta) and
among all Mini-TES spectra from Gusev crater, most
resemble Clovis class rocks (green). Wishstone spectra are
dominated by intermediate plagioclase (labradorite shown
in red; scaled) and clearly match the type example (blue, an
average of four spectra) from earlier in the mission.
Backstay and Adirondack spectra (orange and purple,
respectively) compare favorably to the subset of float rock
spectra shown. Adirondack spectra are dominated by
olivine, as shown by the scaled laboratory-based spectrum
(cyan) that is an average of 60% forsteritic olivine and 35%
forsteritic olivine compositions. Adirondack was examined
early in the mission on the plains [Arvidson et al. 2006a].
E12S33 ARVIDSON ET AL.: MER SPIRIT OVERVIEW AND SELECTED RESULTS
18 of 35
E12S33
two examples that show that water was involved in alter-
ation of crustal materials: the Voltaire outcrops and the
emplacement and modification of volcanic materials in the
Inner Basin. The discovery of evidence for extensive
aqueous alteration is in contrast to what was found by Spirit
during its exploration of the volcanic plains that surround
and embay the Hills. Specifically the plains are dominated
by olivine-bearing basaltic float rocks and wind-blown soils
that are weakly altered relative to the rocks in the Hills
[Arvidson et al., 2006a; Haskin et al., 2005; McSween et al.,
2008; Morris et al., 2008; Ming et al., 2008]. The question
is: Can the evidence for aqueous alteration in the Columbia
Hills also be detected from orbit?
[28] Analysis of Mars Global Surveyor Thermal Emis-
sion Spectrometer (TES) data [e.g., Martinez-Alonso et
al., 2005] and Mars Express OMEGA hyperspectral
imager data (0.4 to 5 mm) [Lichtenberg et al., 2007]
for the Gusev plains are also consistent with the presence
of slightly oxidized basalt sands mixed with and covered
to varying extent with dust. Neither orbital instrument can
resolve landforms within the Columbia Hills since the
pixel size for TES is 
3 km and at best 
350 m for
OMEGA. Thus data from TES and OMEGA cannot be
used to test the hypothesis that Spirit?s observations in the
Columbia Hills of glassy or amorphous deposits, goethite,
hematite, opaline silica, and hydrated sulfates can also be
detected from orbital observations. On the other hand, the
CRISM hyperspectral imager on MRO acquires Full
Resolution targeted (FRT) mode data from 0.4 to 4.0 mm
(545 bands, 18 m/pixel, 
10 km frame width) and is able to
resolve many of the features within the Columbia Hills,
including Husband Hill, El Dorado, and the Inner basin
(Figure 1). In this section of the paper, reduction and
analysis of CRISM observations are detailed and conclu-
sions drawn about what controls the spectral signatures of
the Columbia Hills at the spatial scale sampled by this
instrument.
[29] Several CRISM Full Resolution targeted (FRT)
mode observations were acquired of the Columbia Hills
and surrounding areas while Spirit obtained near simulta-
neous atmospheric optical depth measurements using the
Pancam cameras to image the sun at 0.4 and 0.8 mm and
Mini-TES to obtain temperature profiles of the lower
atmosphere. These observations and dust and ice aerosol
radiative properties based on historical trends from TES
data were used with the DISORT radiative transfer code
[Stamnes et al., 1988] to model CRISM?s spectral radiances.
Atmospheric carbon dioxide, water vapor, carbon monox-
ide, and associated Rayleigh scattering and discrete gas
absorption bands for CO2, H2O, and CO were included in
the computations, along with aerosol scattering and absorp-
tion. Procedures were implemented to retrieve surface
Figure 10. APXS-based factor 1 versus 2 plot from a correspondence analysis run for rocks examined
on Husband Hill. This plane captures 83% of the variance of the APXS data set and shows how rock
analyses relate to one another and which elements provide ??finger prints?? or distinguishing
characteristics for the samples. For example, Independence is enriched in SiO2, K2O, and Al2O3,
whereas Wishstone is enriched in CaO, Na2O, P2O5, and TiO2 relative to the average composition for
rocks examined on Husband Hill. The rock analyses shown in red are for matrix materials (Discourse on
the Descartes Outcrop, Gentile Matrice on the Bourgeoisie Outcrop, and Rue LaPlace on the Haussmann
Outcrop), whereas the blue color corresponds to Chic, a clast in the Bourgeoisie Outcrop. Note that the
matrix materials plot close to the composition average for the rocks, whereas Chic is displaced toward
Wishstone. Assemblee is highly enriched in Cr2O3 and is enriched in SiO2, K2O, and Al2O3 and plots as a
very chemically distinct rock relative to other rocks examined on Husband Hill. APXS data from Ming et
al. [2006] and Ming et al. [2008].
E12S33 ARVIDSON ET AL.: MER SPIRIT OVERVIEW AND SELECTED RESULTS
19 of 35
E12S33
Lambert albedos using DISORT-based simulations and
regressions for each wavelength band between modeled
CRISM radiances and a suite of input gray surface albedos,
following the methodology outlined by Arvidson et al.
[2006b] for modeling OMEGA data over the Opportunity
site. The retrieved surface spectra were examined for
residuals associated with changes in band pass character-
istics relative to prelaunch calibrations and incomplete
removal of atmospheric carbon dioxide and water vapor.
Runs with updated band pass values and atmospheric
conditions relative to historical trends and rover-based
observations allowed convergence on the proper atmospher-
ic model within several iterations.
[30] We concentrate on CRISM FRT00003192_07 ac-
quired over the Columbia Hills and surrounding plains
while Spirit was located at the Low Ridge winter campaign
site (Figure 1). Detailed analysis of the CRISM spectra for
the plains does not show any spectral evidence of the
phyllosilicates, hydrated sulfates, or opaline silica deposits
found elsewhere on the planet using OMEGA and CRISM
observations [e.g., Gendrin et al., 2005; Arvidson et al.,
2005; Poulet et al., 2005; Mustard et al., 2008; Milliken et
al., 2008]. A detailed search was also conducted using the
CRISM data over the Columbia Hills for spectral signatures
from (1) Voltaire and Voltaire-like materials (e.g., water
combination bands at 1.4 and 1.9 mm, metal-OH features
from allophane-like materials between 2 and 2.5 mm); (2)
opaline silica signatures expected from deposits such as the
disturbed soils in Gertrude Weise (e.g., the broad 2.2 mm Si-
OH absorption and the 1.9 mm H2O combination band
detected using Airborne Visible and Infrared Imaging Spec-
trometer (AVIRIS) data in the Kau Desert, Hawaii (Seelos et
al., manuscript in preparation, 2008) and from CRISM data
for selected plains around Valles Marineris [Milliken et al.,
2008]); and (3) multiple H2O combination bands associated
with hydrated sulfates (e.g., water combination bands be-
tween 1 and 2.5 mm). None of these signatures was
detected, even when precisely locating Spirit?s experiment
sites and examining single pixel spectra for these regions.
Instead, the Columbia Hills CRISM spectra are dominated
by the ferric-rich dust (spectrally dominated by nanophase
iron oxides), olivine, and pyroxene, as will be detailed in the
next several paragraphs. The data do show the ubiquitous

3 mm band found across Mars and associated with
adsorbed or absorbed water molecules. The lack of vibra-
tional bands associated with hydrated or hydroxlated min-
erals (i.e., items 1?3 above) is not surprising considering
that the length scale for the Voltaire Outcrop is only 
5 m
Figure 11. Portion of McMurdo Pancam panorama acquired from the Low Ridge winter campaign site
looking to the northeast. The light-toned Tyrone disturbed soils can be seen in the distance, along with
tracks leading to and from Tyrone, and the light-toned soils spilled from the right front wheel cowling
during Spirit?s backward drive from Tyrone to the winter site. Berkner Island is the light-toned soil target
examined in detail by Spirit with its in situ instruments. King George Island is the platy outcrop target
that dips into Low Ridge and that was examined by Spirit, along with the vesicular basalt boulder
Esperanza immediately after leaving the campaign site in the spring. Halley and Bear Island were also
examined in detail as part of the winter campaign. The Troll Outcrop, located in the upper left portion of
the figure, was another location for detailed measurements after leaving the campaign site. Platy outcrops
typified by King George Island wrap around Low Ridge and form a synformal structure. Color image
using the 0.43 for blue, 0.53 for green, and 0.60 mm for red.
E12S33 ARVIDSON ET AL.: MER SPIRIT OVERVIEW AND SELECTED RESULTS
20 of 35
E12S33
by 2.5 m and the rocks are surrounded and partially covered
by wind-blown soil (Figure 4). In addition, the opaline silica
and sulfate soils examined by Spirit were discovered only
after the vehicle exposed them during drives (Figures 11
and 16), while the silica-rich nodular outcrops are small and
heavily contaminated by wind-blown soil. The deposits may
be widespread, but they are hidden from CRISM by a cover
of wind-blown soil. Finally, no signatures expected from
phyllosilicate minerals were detected for any spectra
extracted from the Columbia Hills.
[31] Detailed examination of the CRISM data for the
Columbia Hills regions traversed by Spirit shows that the
spectral end-members are represented by the top of Hus-
band Hill and the upper slopes of Tennessee Valley for the
brightest area and the El Dorado ripple field for the darkest
area (Figures 1, 2, and 19). On the basis of imaging data
acquired while Spirit was at the summit of Husband Hill,
combined with examination of the HiRISE data covering
the summit and Tennessee Valley, it is clear that the upper
portion of Tennessee Valley and the northern portion of
the summit are covered by light-toned ripples that have
migrated from the NW, i.e., up the valley (Figures 2a and
2b) [Sullivan et al., 2008]. Spirit examined one of these
ripples (Cliffhanger) during its summit experiments (Table 2
and Figures 2b and 20). The experiments included measure-
ments of undisturbed and scuffed soils in the ripple. MB
observations of the undisturbed (Hang2) and scuffed (Lands
End) ripple soils indicate in order of decreasing abundance:
pyroxene, nanophase iron oxides, olivine, hematite, and
magnetite (Figure 20 and Table 5). Spirit also conducted
similar experiments on the eastern edge of the El Dorado
Figure 12a. Portion of the McMurdo Pancam panorama
shown in Figure 11 enlarged to show details of Esperanza,
interpreted to be a vesicular basalt boulder. The IDD target
Palma is located on the left side of Esperanza out of site of
the view shown. A second boulder is evident to the upper
left of Esperanza and is interpreted to be a fine-grained rock
shaped by wind into a ventifact. Esperanza is approximately
0.20 m across.
Figure 12b. MI mosaic of Esperanza illustrating the extent to which wind-blown sand and dust has
shaped the surface into a set of sharp curvilinear ridges. Note the sand sitting on the bottom of one of the
vesicles. Frame covers 
6 cm across. Mosaic 2MMA53IOFASORTAFP2936M222F1 with a filtering
and contrast enhancement applied to minimize loss of detail from shadows.
E12S33 ARVIDSON ET AL.: MER SPIRIT OVERVIEW AND SELECTED RESULTS
21 of 35
E12S33
ripple field (Table 2 and Figures 2c and 21), with MB
measurements from the undisturbed surface (Shadow) indi-
cating in order of decreasing abundance: olivine, pyroxene,
magnetite, and nanophase iron oxides (Table 5). Mini-TES
spectra extracted for these surfaces indicate a dominance of
olivine for the El Dorado area based on the bending
vibrational modes evident between 
650 to 400 cm1
and a dust dominance for the Cliffhanger ripple surfaces
based on the overall spectral shape (Figure 22). Pancam
spectra retrieved for Cliffhanger and El Dorado undisturbed
surfaces are similar in overall shape to the CRISM spectra
and consistent with a dominance of nanophase iron oxides
in the visible wavelengths, olivine and pyroxene for longer
wavelengths for the El Dorado spectra, and nanophase iron
oxides and pyroxene for the light-toned ripples that domi-
nate the western summit of Husband Hill and the upper
Tennessee Valley (Figures 7 and 19). The El Dorado spectra
also indicate a thin coating of dust that becomes translucent
or more forward scattering with increasing wavelength, thus
allowing the darker mafic sand to spectrally dominate. This
phenomenon produces a negative slope as documented
analyses of OMEGA data by Lichtenberg et al. [2007] for
the Gusev plains and shown in the laboratory by Fischer
and Pieters [1993] and Johnson and Grundy [2001].
[32] Sullivan et al. [2008] show that El Dorado is a ripple
field that accumulates a thin coating of dust that locally is
removed by dust devils. The light-toned ripples on the top
of Husband Hill, on the other hand, are more coarse grained
and interpreted to be active only during rare high velocity
wind events capable of moving the coarse particles and
otherwise develop a uniform dust cover [Sullivan et al.,
2008]. An optically thick dust cover is certainly consistent
with the observation that the Pancam and Mini-TES spectra
for Cliffhanger are dominated by dust, whereas the MB data
show the presence of nanophase iron oxides (a dust com-
ponent) and pyroxene, i.e., MB has a greater penetration
depth than either Pancam or Mini-TES and can see more of
the underlying pyroxene. Sullivan et al. [2008] also hypoth-
esize on the basis of the presence of angular grains (i.e., not
Figure 13. Mini-TES spectra for Esperanza and 12 other
vesicular rocks from Low Ridge are shown in black (red is
the average) and overlain with the pyroxene-rich Bounce
rock spectrum (in green; this rock is a pyroxenite from
Meridiani Planum and was observed by Opportunity) to
demonstrate a spectral dominance by pyroxene. Pyroxene-
rich rocks represent another spectral class of rock observed
by Mini-TES that includes those dominated by glass or
amorphous phases as represented by the average Assemblee
spectrum (blue), the plagioclase-dominated Wishstone
spectrum (purple), and the olivine-dominated spectrum
represented by Adirondack basalt (magenta).
Figure 14. Microscopic Imager view ofKingGeorge Island,
a friable, platy deposit (see Figure 11) of round, uniformly sand
sized grains. Average grain size is 
1 mm. Mosaic covers

6 cm in width. Wind-blown soil can be seen surrounding the
rock at the top and bottom of the mosaic. The composition of
this outcrop is mainly SiO2, followed in relative abundance
by FeO, MgO, and Al2O3 (Table 6). The iron mineralogy is
dominated by hematite (43%; Table 5 and Figure 15),
implying alteration under oxidizing conditions to remove iron
from ferromagnesian silicates to form hematite. The dark
lines are MI poker shadows. Mosaic generated from frames
2M217894337IFFAS20P2956M2F1, 2M217894630IF-
FAS20P2956M2F1, 2M217894945IFFAS20P2956M2F1,
and 2M217895270IFFAS20P2956M2F1 acquired on sol
1031.
E12S33 ARVIDSON ET AL.: MER SPIRIT OVERVIEW AND SELECTED RESULTS
22 of 35
E12S33
Figure 15. MB-based correspondence analysis of all rocks in the Inner Basin showing hematite
enrichment for the friable, platy outcrops exposed on the floor of Eastern Valley and Low and Mitcheltree
Ridges. Numbers refer to the fraction of the iron that is contained within hematite. Montalva is
stratigraphically below the Riquelme Outcrop at Troll. A series of measurements were made on different
sections of Halley and are shown, along with King George Island. Hillary is a rock from the summit of
Husband Hill with a fairly large amount of nanophase iron oxide, magnetite, and hematite [Morris et al.,
2008]. Factors 1 and 3 are used because this projection shows the hematite trend better than other planes.
Factors 1 and 3 capture 67% of the variance in the data set.
Figure 16. Pancam false color mosaic showing the Nancy Warren, Innocent Bystander, and Norma
Luker targets in detail. Innocent Bystander and Norma Luker represent rocks broken apart by wiggling
(using the operative azimuthal actuator) and dragging Spirit?s right front wheel over them. Note the debris
field produced when the rover backed up after doing its crushing. The dark rock labeled vesicular basalt
is typical of numerous clasts and boulders shown to be of basaltic composition. The Pancam frames used
were 2P235913627ESFAU37P2378L2M1, 2P235913690ESFAU37P2378L5M1, 2P235913733ES-
FAU37P2378L7M1, 2P235913941ESFAU37P2378L2M1, 2P235914094ESFAU37P2378L5M1, and
2P235914221ESFAU37P2378L7M1.
E12S33 ARVIDSON ET AL.: MER SPIRIT OVERVIEW AND SELECTED RESULTS
23 of 35
E12S33
Figure 17. Factors 1 versus 2 plot from correspondence analysis applied to APXS data for Inner Basin
targets. Note that the SiO2 and SO3 vectors plot toward the top of the diagram and that key targets follow
these vectors and show progressive enrichments in both SiO2 and SO3, although the targets split into
SiO2- and SO3-dominated trends. Numbers and labels in yellow are for SO3 concentrations, whereas
those in green are for SiO2 concentrations. Arad (Samra) is the disturbed soil with the highest abundance
of SO3 (35%) and Kenosha Comets in the Gertrude Weise disturbed soil has the highest abundance of
SiO2. This projection captures 87% of the variance of the multivariate data set.
E12S33 ARVIDSON ET AL.: MER SPIRIT OVERVIEW AND SELECTED RESULTS
24 of 35
E12S33
Figure 18. Color image from the Kau Desert, Hawaii, showing differentially eroded accretionary lapilli
basaltic ash deposits draped over vesicular basalt flow outcrops and boulders. Rock hammer for scale.
E12S33 ARVIDSON ET AL.: MER SPIRIT OVERVIEW AND SELECTED RESULTS
25 of 35
E12S33
rounded by extensive erosion and transport) that the Cliff-
hanger ripples contain a significant contribution from local
outcrops in Tennessee Valley. In fact, correspondence anal-
ysis applied to elemental abundances determined for soils
by Spirit shows that the Cliffhanger ripples have a chemical
affinity for Wishstone rocks (Figure 23). As noted in
section 3 of this paper, Wishstone materials dominate the
rock population on Husband Hill on the basis of Mini-TES
spectra. Wishstone (and Watchtower, its altered equivalent
[Ming et al., 2008]) will add feldspar (Figure 9, from Mini-
TES data) to the light-toned ripples in the Tennessee Valley
and the northwestern side of the Husband Hill summit.
Feldspar would be spectrally neutral and thus not uniquely
detectable in the Pancam and CRISM wavelength intervals.
Thus, the orbital and surface observations combined pro-
vide a self-consistent picture of what is controlling the
spectral variety at the 18 m/pixel scale relevant to the
CRISM measurements, i.e., regional-scale aeolian mixing
of dust and sand with local materials.
[33] Examination of CRISM spectra for the Columbia
Hills shows a smooth variation in albedo and shape consis-
tent with mixing between the end-members described in the
last paragraphs and shown in Figure 19. This is not
Figure 19. CRISM spectra retrieved from FRT00003192?
07 for light-toned ripples on Husband Hill and Tennessee
Valley and the dark El Dorado ripple field, overlain Pancam
spectra retrieved for undisturbed surfaces on the Cliffhanger
and El Dorado ripples. Data are shown between 0.4 and
2.5 mm only to emphasize comparisons with Pancam data,
although retrievals extended to 4.0 mm. The spectral data are
consistent with control by iron-bearing minerals, with strong
ferric absorption edges shortward of 0.8 mm, broad
absorptions between 0.8 and 1.5 mm due to olivine and
pyroxene, and a negative slope longward of 0.8 mm for the
El Dorado spectrum due to a thin dust cover that becomes
translucent at long wavelengths to reveal the underlying
mafic sand signature. The upturn for the longest Pancam
wavelength for the Cliffhanger ripple is interpreted to be a
consequence of a large dust component for this surface
(
10 cm wide patch) as compared to the CRISM data (3 
3 pixel averages at 18 m/pixel) for the two spectra. Gap in
CRISM data 
0.7 mm is a nonrecoverable portion of the
spectrum where two detectors join. Gap just longward of
1 mm is the join between the S and L CRISM detectors
[Murchie et al., 2007]. Pancam spectra extracted from
scenes quoted in Figure 7 caption.
Figure 20. Hazcam frame covering Cliffhanger ripple
scuff experiments on the summit of Husband Hill. Hang2 is
the location for the MB and APXS undisturbed surface
measurements. View is looking toward the northwest into
the Tennessee Valley and shows the set of light-toned
ripples that have migrated toward the summit. Frame
2F180078494RSLAEM9P1214L0MZ acquired on sol 605.
Figure 21. Hazcam frame covering El Dorado scuff
experiments and looking back toward Husband Hill.
Shadow is the location for the MB undisturbed surface
measurements. Frame 2F189393623RSLAL00P1121L0MZ
acquired on sol 710.
E12S33 ARVIDSON ET AL.: MER SPIRIT OVERVIEW AND SELECTED RESULTS
26 of 35
E12S33
Figure 22. Mini-TES spectra are shown for Cliffhanger and El Dorado ripple surfaces, together with
spectra from a visually bright and dusty hollow on the rim of Bonneville Crater (Figure 1) and dark
ripples within that crater. The dark ripple fields have spectra controlled by olivine (particularly the long-
wavelength region, compare to Figure 9 for Adirondack). The light-toned ripple surfaces are dominated
by dust signatures.
E12S33 ARVIDSON ET AL.: MER SPIRIT OVERVIEW AND SELECTED RESULTS
27 of 35
E12S33
Figure 23. APXS-based correspondence analysis for all soils examined by Spirit. The Wishstone float
rock analysis is also shown. The appended ??_d?? means that the soil has been disturbed or excavated by
Spirit?s wheels. The range of natural soil surfaces varies from El Dorado (basaltic sand, with both
disturbed and undisturbed samples shown) to Lambert Whymper (basaltic sand and dust). Eileen Dean is
located at an extreme position because of its abundance of Cr2O3 and MgO and as noted in the text has
been altered. Cliffhanger and Pequod Doubloon are soils that have the addition of local Wishstone
materials. Boroughs and Big Hole are subsurface trench soil analyses showing an enrichment in SO3.
Data from Gellert et al. [2006] and Ming et al. [2008]. Captures 78% of the variance of the data set.
E12S33 ARVIDSON ET AL.: MER SPIRIT OVERVIEW AND SELECTED RESULTS
28 of 35
E12S33
Figure 24. MB-based correspondence analysis showing trends from basaltic sand with a small amount
of dust indicated by high abundance of olivine, pyroxene, and magnetite (e.g., El Dorado ripple field) to
basaltic sand with a significant amount of dust delineated by nanophase iron oxides, e.g., Cliffhanger
Hang2 on the Cliffhanger light-toned ripple field. Numbers represent Fe+3/Fetotal ratios. Chromite has not
been included in the correspondence analysis since this mineral was only found in the Assemblee rock
[Morris et al., 2008]. The _d for soils refers to scuffed soils or soils disturbed by Spirit?s wheels during
drives. Cliffhanger Hang2 is the undisturbed surface for Cliffhanger, whereas Cliffhanger LandsEnd_d
corresponds to the measurement within the scuff zone. Eileen Dean is a highly altered soil exposed by the
rover?s wheels, as noted in the text. The factors 1 versus 2 plane projection accounts for 78% of the
multidimensional data set variance.
E12S33 ARVIDSON ET AL.: MER SPIRIT OVERVIEW AND SELECTED RESULTS
29 of 35
E12S33
surprising considering that Spirit has mainly encountered
wind-blown soils on West Spur, Husband Hill, and the Inner
Basin, i.e., the mineralogical variations and aqueous alter-
ation associated with the various rock types encountered by
Spirit are not evident because the rock areal footprint is
small relative to soil and dust covers. A spectral dominance
by iron bearing minerals is also evident when comparing the
patterns for MB observations for soils examined by Spirit to
the end-members found in the CRISM spectra (Figures 19
and 24). The Cliffhanger undisturbed surface MB observa-
tion (Hang2) has the highest ratio of Fe+3/Fetotal (0.45) and
the highest abundance of nanophase iron oxides of any soil
(including the plains soils) whereas El Dorado (0.17) has
the lowest ratio and highest proportion of olivine and
pyroxene (Table 5). The vast majority of soils fall on a
mixing line between El Dorado and Hang2 with olivine and
pyroxene dominating the former and nanophase iron oxide
in much higher abundance in the latter. Magnetite pulls the
soil data off a simple mixing line, as expected, with an
extreme case defined by the disturbed soil, Eileen Dean,
found close to Nancy Warren (Figure 16). But, this material
would not be evident from orbit because it was found in
one of Spirit?s wheel tracks. Eileen Dean, with its unusually
high concentration of magnetite, MgO, Cr2O3, Zn, Ni, and
Cl, unusually low NaO and Al2O3, and the presence in
Mini-TES of a 
6 mm water bending mode vibration, is
unusual in many respects and probably represents yet
another type of aqueous alteration not observable from
orbit because of the ubiquitous cover of wind-blown sand
[Morris et al., 2008; Ming et al., 2008].
6. Conclusions and Implications
[34] Spirit has shown extensive evidence for interaction
of water and crustal materials over its 
1500 sols of
exploration. This includes Voltaire, altered impact or volca-
nic rocks on Husband Hill, and the formation of volcani-
clastic materials with hematite, sulfate, and opaline silica
enrichment produced by aqueous processes in and around
Home Plate in the Inner Basin. On the other hand, orbital
spectral observations by the CRISM hyperspectral imager
(0.4?4.0 mm) are dominated by the presence of iron-bearing
phases. The discrepancy between extensive evidence for
aqueous alteration discovered by Spirit and the dominance
of iron-bearing minerals from orbit is resolved when it is
realized that windblown deposits dominate the traverse sites
by areal extent and that the evidence for alteration observed
by Spirit occur in outcrops that are too small and/or mostly
buried to be resolved in the 18 m/pixel orbital data.
OMEGA and CRISM observations show the presence of
phyllosilicates, hydrated sulfates, and opaline silica else-
where on Mars where these deposits are exposed on spatial
length scales properly sampled by the spatial footprints for
these sensors. The fact that the Columbia Hills shows
evidence for extensive aqueous alteration when examined
in detail over much smaller length scales (i.e., from Spirit)
Figure A1. Overview of Spirit?s traverses. Image covers 
1900 m in width.
E12S33 ARVIDSON ET AL.: MER SPIRIT OVERVIEW AND SELECTED RESULTS
30 of 35
E12S33
bodes well for continued and detailed observations by future
rover missions that are focused on ??following the water??
and evaluating whether or not the planet was once habitable.
We suggest that older terrains on Mars are likely to have
been ubiquitously altered by aqueous processes, increasing
the likelihood that the planet was once habitable and that the
evidence has been preserved.
Appendix A
[35] In this Appendix we provide detailed traverse informa-
tion for Spirit, overlain on HiRISE image PSP_001513_1655_
red, projected to an equirectangular map base. To precisely
locate the rover on any given sol the ensemble of rover
observations was used by us to locate the position on the
HiRISE image using a combination of nearby and horizon
landmarks. The locations were cross-checked against a
bundle-adjusted set of locations derived from stereo image
analysis by Li et al. [2008]. The presentation includes an
overview map (Figure A1) showing the traverses, together
with boxes that show enlargements in Figures A2?A8. Sols,
sites, and key terrain features are shown on the enlargements,
although the high density of experiment sites in EasternValley
Figure A2. Traverses are shown across West Spur.
Figure A3. Traverses are shown across the northern portion of Spirit?s travels on Husband Hill.
E12S33 ARVIDSON ET AL.: MER SPIRIT OVERVIEW AND SELECTED RESULTS
31 of 35
E12S33
(Figure A8) precludes delineation of each location. These
overlays should be used in combination with Tables 2 and 3,
Figure 3, and the description of operations provided in the
main body of the paper to understand what Spirit has
accomplished. Note that the box labels refer to relative
locations of the boxes on Husband Hill and not to absolute
cardinal positions. The detailed traverses and measurements
conducted in Eastern valley are not possible to show in detail
on a HiRISE base map.
Figure A4. Traverses are shown across the southwestern portion of Spirit?s travels on Husband Hill.
Figure A5. Traverses are shown across the southeastern portion of Spirit?s travels on Husband Hill.
E12S33 ARVIDSON ET AL.: MER SPIRIT OVERVIEW AND SELECTED RESULTS
32 of 35
E12S33
Figure A6. Spirit?s traverses are shown across the Haskin
Ridge and areas to the south.
Figure A7. Spirit?s traverses are shown across a portion of
the El Dorado area and the northern portion of the Inner
Basin.
E12S33 ARVIDSON ET AL.: MER SPIRIT OVERVIEW AND SELECTED RESULTS
33 of 35
E12S33
[36] Acknowledgments. We thank the capable team of engineers and
scientists who made the Spirit mission possible, and we thank NASA for its
support of our endeavors and the HiRISE, CTX, and CRISM teams who
worked to acquire, process, and release the orbital data used in this study.
We also thank Bethany Ehlmann and Horton Newsom for thoughtful
reviews. C. S. specifically acknowledges support by an appointment to
the NASA Postdoctoral Program at the Johnson Space Center, administered
by Oak Ridge Associated Universities through a contract with NASA.
References
Arvidson, R. E., F. Poulet, J.-P. Bibring, M. Wolff, A. Gendrin, R. V.
Morris, J. J. Freeman, Y. Langevin, N. Mangold, and G. Bellucci
(2005), Spectral reflectance and morphologic correlations in eastern Terra
Meridiani, Mars, Science, 307(5715), 1591 ? 1594, doi:10.1126/
science.1109509.
Arvidson, R. E., et al. (2006a), Overview of the Spirit Mars Exploration
Rover Mission to Gusev Crater: Landing site to Backstay Rock in the
Columbia Hills, J. Geophys. Res., 111, E02S01, doi:10.1029/
2005JE002499.
Arvidson, R. E., et al. (2006b), Nature and origin of the hematite-bearing
plains of Terra Meridiani based on analysis of orbital and Mars Explora-
tion Rover data sets, J. Geophys. Res., 111, E12S08, doi:10.1029/
2006JE002728.
Burns, R. (1993), Origin of electronic spectra of minerals in the visible
to near-infrared region, in Remote Geochemical Analysis: Elemental
and Mineralogical Composition, edited by C. Pieters and P. Englert,
pp. 3?29, Cambridge Univ. Press, New York.
Clark, B. C., et al. (2007), Evidence for montmorillonite or its composi-
tional equivalent in Columbia Hills, Mars, J. Geophys. Res., 112,
E06S01, doi:10.1029/2006JE002756.
Farrand, W. H., J. F. Bell III, J. R. Johnson, S. W. Squyres, J. Soderblom,
and D. W. Ming (2006), Spectral variability among rocks in visible
and near infrared multispectral Pancam data collected at Gusev Crater:
Examinations using spectral mixture analysis and related techniques,
J. Geophys. Res., 111, E02S15, doi:10.1029/2005JE002495.
Fischer, E. M., and C. M. Pieters (1993), The continuum slope of Mars:
Bidirectional reflectance investigations and applications to Olympus
Mons, Icarus, 102, 185?202, doi:10.1006/icar.1993.1043.
Gellert, R., et al. (2006), Alpha Particle X-Ray Spectrometer (APXS):
Results from Gusev crater and calibration report, J. Geophys. Res., 111,
E02S05, doi:10.1029/2005JE002555.
Gendrin, A., et al. (2005), Sulfates in Martian layered terrains: The OMEGA/
Mars express view, Science, 307(5715), 1587 ? 1591, doi:10.1126/
science.1109087.
Haskin, L. A., et al. (2005), Water alteration of rocks and soils on Mars at
the Spirit rover site in Gusev Crater, Nat., 436, 66?69, doi:10.1038/
nature03640.
Johnson, J. R., and W. M. Grundy (2001), Visible/near-infrared spectra and
two-layer modeling of palagonite-coated basalts, Geophys. Res. Lett., 28,
2101?2104, doi:10.1029/2000GL012669.
Johnson, J. R., J. F. Bell III, E. Cloutis, M. Staid, W. H. Farrand, T. McCoy,
M. Rice, A. Wang, and A. Yen (2007), Mineralogic constraints on sulfur-
rich soils from Pancam spectra at Gusev crater, Mars, Geophys. Res. Lett.,
34, L13202, doi:10.1029/2007GL029894.
Kurtz, A. C., L. A. Derry, and O. A. Chadwick (2002), Germanium-silicon
fractionation in the weathering environment, Geochem. Cosmochim.
Acta, 66, 1525?1537.
Lewis, K., O. Aharonson, and M. Schmidt (2007), Stratigraphy and struc-
ture of Inner Basin outcrops in the Columbia Hills from the Spirit rover,
paper presented at 38th Lunar and Planetary Science Conference, Lunar
and Planet. Inst., League City, Tex.
Lewis, K. W., O. Aharonson, J. P. Grotzinger, S. W. Squyres, and M. E.
Schmidt (2008), Structure and stratigraphy of Home Plate from the Spirit
Mars Exploration Rover, J. Geophys. Res., doi:10.1029/2007JE003025,
in press.
Li, R., et al. (2008), Characterization of traverse slippage experienced by
Spirit rover on Husband Hill at Gusev Crater, J. Geophys. Res.,
doi:10.1029/2008JE003097, in press.
Lichtenberg, K. A., et al. (2007), Coordinated analyses of orbital and Spirit
Rover data to characterize surface materials on the cratered plains of Gusev
Crater, Mars, J. Geophys. Res., 112, E12S90, doi:10.1029/2006JE002850.
Madsen, M. B., et al. (2008), Overview of the Magnetic Properties Experi-
ments on the Mars Exploration Rovers, J. Geophys. Res., doi:10.1029/
2008JE003098, in press.
Malin, M. C., D. Dzurisin, and R. P. Sharp (1983), Stripping of Keanakakoi
tephra on Kilauea Volcano, Hawaii, Geol. Soc. Am. Bull., 94, 1148?
1158, doi:10.1130/0016-7606(1983)94<1148:SOKTOK>2.0.CO;2.
Figure A8. Spirit?s traverses are shown across Home Plate and surrounding areas.
E12S33 ARVIDSON ET AL.: MER SPIRIT OVERVIEW AND SELECTED RESULTS
34 of 35
E12S33
Malin, M. C., et al. (2007), Context camera investigation on board the Mars
Reconnaissance Orbiter, J. Geophys. Res., 112, E05S04, doi:10.1029/
2006JE002808.
Martinez-Alonso, S., B. M. Jakosky, M. T. Mellon, and N. E. Putzig (2005),
A volcanic interpretation of Gusev crater surface materials from thermo-
physical, spectral, and morphological evidence, J. Geophys. Res., 110,
E01003, doi:10.1029/2004JE002327.
McCoy, T. J., et al. (2008), Structure, stratigraphy, and origin of Husband
Hill, Columbia Hills, Gusev Crater, Mars, J. Geophys. Res., 113, E06S03,
doi:10.1029/2007JE003041.
McEwen, A. S., et al. (2007), Mars Reconnaissance Orbiter?s High Resolu-
tion Imaging Science Experiment (HiRISE), J. Geophys. Res., 112,
E05S02, doi:10.1029/2005JE002605.
McSween, H. Y., et al. (2006), Alkaline volcanic rocks from the Columbia
Hills, Gusev crater, Mars, J. Geophys. Res., 111, E09S91, doi:10.1029/
2006JE002698.
McSween, H. Y., et al. (2008), Mineralogy of volcanic rocks in Gusev
Crater, Mars: Reconciling Mo?ssbauer, Alpha Particle X-Ray Spectro-
meter, and Miniature Thermal Emission Spectrometer spectra, J. Geo-
phys. Res., 113, E06S04, doi:10.1029/2007JE002970.
Milliken, R. E., et al. (2008), Opaline silica in young deposits on Mars,
Geology, in press.
Ming, D. W., et al. (2006), Geochemical and mineralogical indicators for
aqueous processes in the Columbia Hills of Gusev Crater, Mars, J. Geo-
phys. Res., 111, E02S12, doi:10.1029/2005JE002560.
Ming, D. W., et al. (2008), Geochemical properties of rocks and soils in
Gusev Crater, Mars: Results of the Alpha Particle X-Ray Spectrometer
from Cumberland Ridge to Home Plate, J. Geophys. Res., doi:10.1029/
2008JE003195, in press.
Morris, R. V., et al. (2000), Mineralogy, composition, and alteration of Mars
Pathfinder rocks and soils: Evidence from multispectral, elemental, and
magnetic data on terrestrial analogue, SNC meteorite, and Pathfinder sam-
ples, J. Geophys. Res., 105(E1), 1757?1817, doi:10.1029/1999JE001059.
Morris, R. V., et al. (2006), Mo?ssbauer mineralogy of rock, soil, and dust at
Gusev Crater, Mars: Spirit?s journey through weakly altered olivine basalt
on the Plains and pervasively altered basalt in the Columbia Hills,
J. Geophys. Res., 111, E02S13, doi:10.1029/2005JE002584.
Morris, R. V., et al. (2008), Iron mineralogy and aqueous alteration from
Husband Hill through Home Plate at Gusev Crater, Mars, from the Mo?ss-
bauer Instrument on the Spirit Mars Exploration Rover, J. Geophys. Res.,
doi:10.1029/2008JE003201, in press.
Murchie, S. L., et al. (2007), Compact Reconnaissance Imaging Spectro-
meter for Mars (CRISM) on Mars Reconnaissance Orbiter (MRO),
J. Geophys. Res., 112, E05S03, doi:10.1029/2006JE002682.
Mustard, J. F., et al. (2008), Hydrated silicate minerals on Mars observed
by the CRISM instrument on MRO, Nat., 454, 305?309, doi:10.1038/
nature07097.
Parfitt, R. L., and R. J. Furkert (1980), Identification and structure of two
types of allophane from volcanic ash soils and tephra, Clays Clay Miner.,
28, 328?334, doi:10.1346/CCMN.1980.0280502.
Parke, S. (1974), Glasses, in The Infrared Spectra of Minerals, edited by
V. C. Farmer, pp. 483?514, Mineral. Soc. of G. B. and Ireland, Twickenham,
U. K.
Poulet, F., J.-P. Bibring, J. F. Mustard, A. Gendrin, N. Mangold, Y. Langevin,
R. E. Arvidson, B. Gondet, and C. Gomez (2005), Phyllosilicates on Mars
and implications for the early Mars History, Nat., 438, 623 ? 627,
doi:10.1038/nature04274.
Rice, M., J. Bell, A. Wang, and E. Cloutis (2008), VIS-NIR spectral
characterization of Si-rich deposits at Gusev Crater, Mars, paper
presented at 39th Lunar and Planetary Science Conference, Lunar and
Planet. Inst., League City, Tex.
Ruff, S. W., P. R. Christensen, D. L. Blaney, W. H. Farrand, J. R. Johnson,
J. R. Michalski, J. E. Moersch, S. P. Wright, and S. W. Squyres (2006),
The rocks of Gusev Crater as viewed by the Mini-TES instrument,
J. Geophys. Res., 111(E12), E12S18, doi:10.1029/2006JE002747.
Schiffman, P., R. Zierenberg, N. Marks, J. Bishop, and M. Dyer (2006),
Acid-fog deposition at Kilauea volcano: A possible mechanism for the
formation of siliceous rock coatings on Mars, Geol. Soc. Am. Bull., 34,
921?924.
Schmidt, M., et al. (2008), Hydrothermal origin of halogens at Home Plate,
Gusev Crater, J. Geophys. Res., 113, E06S12, doi:10.1029/2007JE003027.
Squyres, S. W., et al. (2003), Athena Mars rover science investigation,
J. Geophys. Res., 108(E12), 8062, doi:10.1029/2003JE002121.
Squyres, S. W., et al. (2006), Rocks of the Columbia Hills, J. Geophys.
Res., 111, E02S11, doi:10.1029/2005JE002562.
Squyres, S. W., et al. (2007), Pyroclastic activity at Home Plate in Gusev
Crater, Mars, Science, 316, 738?742, doi:10.1126/science.1139045.
Squyres, S. W., et al. (2008), Detection of silica-rich deposits on Mars,
Science, 320, 1063?1067, doi:10.1126/science.1155429.
Stamnes, K., S. Tsay, W. Wiscombe, and K. Jayaweera (1988), Numerically
stable algorithm for discrete-ordinate-method radiative transfer in multi-
ple scattering and emitting layered media, Appl. Opt., 27, 2502?2509.
Sullivan, R., R. Arvidson, J. F. Bell III, M. Golombek, R. Greeley,
K. Herkenhoff, J. Johnson, S. Thompson, P. Whelley, and J. Wray (2008),
Wind-driven particle mobility on Mars: Insights from MER observations
at ??El Dorado?? and surroundings at Gusev Crater, J. Geophys. Res., 113,
E06S07, doi:10.1029/2008JE003101.
Wang, A., et al. (2008), Light-toned salty soils and co-existing Si-rich
species discovered by the Mars Exploration Rover Spirit in Columbia
Hills, J. Geophys. Res., doi:10.1029/2008JE003126, in press.
Yen, A. S., et al. (2008), Hydrothermal processes at Gusev Crater: An
evaluation of Paso Robles class soils, J. Geophys. Res., 113, E06S10,
doi:10.1029/2007JE002978.

R. E. Arvidson, R. Greenberger, E. A. Guinness, A.Wang, and S.Wiseman,
Department of Earth and Planetary Sciences, Washington University, 1
Brookings Drive, St. Louis, MO 63130, USA. (arvidson@rsmail.wustl.edu)
J. F. Bell III, S. W. Squyres, and R. J. Sullivan, Department of Astronomy,
Cornell University, 610 Space Sciences Building, Ithaca, NY 14853, USA.
N. A. Cabrol, NASAAmes/SETI Institute, Moffett Field, CA 94035, USA.
B. C. Clark, Lockheed Martin Corporation, 12999 West Deer Creek
Canyon Road, Littleton, CO 80125, USA.
L. S. Crumpler, New Mexico Museum of Natural History and Science,
1801 Mountain Road NW, Albuquerque, NM 87104, USA.
W. H. Farrand, Space Science Institute, 4750 Walnut Street, Suite 205,
Boulder, CO 80301, USA.
R. Gellert, Department of Physics, University of Guelph, MacNaughton
Building, Gordon Street, Guelph, ON, Canada N1G 2W1.
W. Goetz, Max Planck Institute for Sonnensystemforschung, 2 Max-
Planck-Strasse 2, D-37191 Katlenburg-Lindau, Germany.
J. A. Grant, Center for Earth and Planetary Studies, National Air and
Space Museum, Smithsonian Institution, P.O. Box 37012, Washington, DC
20013, USA.
K. E. Herkenhoff and J. R. Johnson, U. S. Geological Survey, 2255 North
Gemini Drive, Flagstaff, AZ 86001, USA.
J. A. Hurowitz and A. S. Yen, Jet Propulsion Laboratory, California
Institute of Technology, 4800 Oak Grove Drive, Pasadena, CA 91109,
USA.
G. Klingelho?fer, Institut fu?r Anorganische und Analytische Chemie,
Johannes Gutenberg-Universita?t, Duesbergweg 10-14, D-55099 Mainz,
Germany.
K. W. Lewis, California Institute of Technology, 1200 East California
Boulevard, Pasadena, CA 91125, USA.
R. Li and M. Schmidt, Department of Civil and Environmental
Engineering and Geodetic Science, Ohio State University, 470 Hitchcock
Hall, 2070 Neil Avenue, Columbus, OH 43210, USA.
M. B. Madsen, Niels Bohr Institute, University of Copenhagen,
Blegdamsvej 17, DK-2100 Copenhagen, Denmark.
T. J. McCoy, Department of Mineral Sciences, Smithsonian Institution,
P.O. Box 37012, Washington, DC 20013, USA.
S. M. McLennan, Department of Geosciences, State University of New
York, 255 Earth and Space Sciences Building, Stony Brook, NY 11794,
USA.
H. Y. McSween and J. Moersch, Department of Earth and Planetary
Sciences, University of Tennessee, 306 Earth and Planetary Sciences
Building, Knoxville, TN 37996, USA.
D. W. Ming, R. V. Morris, and C. Schro?der, Johnson Space Center,
NASA, 2101 NASA Parkway, Houston, TX 77058, USA.
S. L. Murchie, Applied Physics Laboratory, Johns Hopkins University,
11100 Johns Hopkins Road, Laurel, MD 20723, USA.
S. W. Ruff, School of Earth and Space Exploration, Arizona State
University, P.O. Box 871404, Tempe, AZ 85287, USA.
E12S33 ARVIDSON ET AL.: MER SPIRIT OVERVIEW AND SELECTED RESULTS
35 of 35
E12S33