Opportunity Mars Rover mission: Overview and selected results from Purgatory ripple to traverses to Endeavour crater



[1] Opportunity has been traversing the Meridiani plains since 25 January 2004 (sol 1), acquiring numerous observations of the atmosphere, soils, and rocks. This paper provides an overview of key discoveries between sols 511 and 2300, complementing earlier papers covering results from the initial phases of the mission. Key new results include (1) atmospheric argon measurements that demonstrate the importance of atmospheric transport to and from the winter carbon dioxide polar ice caps; (2) observations showing that aeolian ripples covering the plains were generated by easterly winds during an epoch with enhanced Hadley cell circulation; (3) the discovery and characterization of cobbles and boulders that include iron and stony-iron meteorites and Martian impact ejecta; (4) measurements of wall rock strata within Erebus and Victoria craters that provide compelling evidence of formation by aeolian sand deposition, with local reworking within ephemeral lakes; (5) determination that the stratigraphy exposed in the walls of Victoria and Endurance craters show an enrichment of chlorine and depletion of magnesium and sulfur with increasing depth. This result implies that regional-scale aqueous alteration took place before formation of these craters. Most recently, Opportunity has been traversing toward the ancient Endeavour crater. Orbital data show that clay minerals are exposed on its rim. Hydrated sulfate minerals are exposed in plains rocks adjacent to the rim, unlike the surfaces of plains outcrops observed thus far by Opportunity. With continued mechanical health, Opportunity will reach terrains on and around Endeavour's rim that will be markedly different from anything examined to date.

1. Introduction

[2] The Mars Exploration Rover (MER) Opportunity touched down on the Meridiani plains on 25 January 2004 (Figures 12). Since landing, Opportunity has conducted numerous traverses and made extensive measurements with its Athena science payload (Table 1), including examination of impact crater ejecta deposits, rims, and walls to access and characterize stratigraphic rock sections within the Burns formation [e.g., Squyres and Knoll, 2005], detailed examination of a variety of cobbles and boulders exposed on the surface, and characterization of the aeolian ripples that partially cover plains outcrops. In addition, numerous atmospheric opacity and cloud measurements have been acquired using Pancam and Navcam, and the Alpha Particle X-Ray Spectrometer (APXS) has been used to monitor atmospheric argon mixing ratios.

Figure 1.

Geologic map for the southern portion of the Meridiani Planum layered sedimentary rocks (ph, hematite-bearing plains) and the Noachian-aged dissected cratered terrain (dct). Miyamoto is a partially buried impact basin that predates deposition of the layered sedimentary sequence. Endeavour and Iazu are Noachian craters partially buried by the layered materials, as is the channel system mapped as “ch.” The ph unit unconformably overlies the dct and ch units and is interpreted from impact crater densities to have been emplaced in late Noachian to early Hesperian times [Arvidson et al., 2006]. The impact event that produced the crater Bopolu penetrated below the layered materials (ph) and into the underlying Noachian cratered terrain materials (dct). These older materials are exposed on the crater floor and ejecta deposits. Opportunity's ∼22 km (as of 27 July 2010) of traverses are overlain as a black line. A THEMIS daytime IR mosaic was used as a map base. White box delineates area shown in Figure 11.

Figure 2.

Portion of a CTX image P15_006847_1770_XN_03S005W_080111 covering Opportunity's traverses and immediate surroundings. Eagle, Endurance, Erebus, Victoria, Raleigh, and Concepción craters are shown, along with the traverse direction to Endeavour, located about 11 km to the southeast of the sol 2300 location. Block Island is an iron meteorite discovered south of Victoria. Key sols are shown, including the location of the Purgatory ripple, where the previous overview paper [Squyres et al., 2006] ended its summary of operations and science highlights. Bright areas correspond to regions with extensive outcrop, whereas darker areas are widely covered by aeolian ripples.

Table 1. Opportunity's Payload Elementsa
InstrumentKey Parameters
  • a

    Magnets were also included on the spacecraft but not described in this table.

Mast-Mounted Science Instruments
Pancam: Panoramic CameraMultispectral imager (∼400 to 1000 nm) with stereoscopic capability; 0.28 mrad IFOV; 16.8 deg by 16.8°FOV. Stereo baseline separation of 30 cm. External calibration target on rover deck.
Mini-TES: Thermal Emission SpectrometerEmission spectra (5 to 29 μm, 10 cm−1 resolution) with 8 or 20 mrad FOV. Internal and external blackbody calibration targets. Instrument put in “stand down” mode on sol 2257 after failing to respond to commands.
Instrument Deployment Device (IDD)-Mounted In Situ Package
APXS: Alpha Particle X-Ray Spectrometer244Cm alpha particle sources, and x-ray detectors, 3.8 cm FOV.
MB: Mössbauer Spectrometer57Fe spectrometer in backscatter mode; 57Co/Rh source and Si-PIN diode detectors; field of view approximately 1.5 cm.
MI: Microscopic Imager31 μm/pixel monochromatic imager (1024 × 1024) with 2 mm depth of field.
RAT: Rock Abrasion ToolTool capable of brushing surfaces and grinding 5 mm deep by 4.5 cm wide surface on rocks.
Engineering Cameras
Navigation Cameras (Navcam)Mast-mounted panchromatic stereoscopic imaging system with 0.77 mrad IFOV; 45°FOV, and 20 cm stereo baseline separation.
Hazard Avoidance Cameras (Hazcam)Front and rear-looking panchromatic stereoscopic imaging systems with 2 mrad IFOV; 123°FOV, 10 cm stereo baseline separation.

[3] The purpose of this paper is to summarize operations and present selected scientific highlights from the time Opportunity left the Purgatory (unless otherwise noted, names for features used in this paper are informal) aeolian ripple on sol 511 (1 July 2005) to the first relatively high spatial resolution views of the Endeavour crater rim on sol 2300 (13 July 2010; Figure 1 and Table 2). The paper also includes a synthesis of orbital and rover-based data for areas along and close to Opportunity's traverses for interpretations of material properties, morphology, and geologic histories on both local and regional scales. This overview is meant to complement papers that provide detailed findings from Opportunity's measurements that are included as the fourth set of Mars Exploration Rover papers published in the Journal of Geophysical Research and also published elsewhere over the past several years. For reference, Squyres et al. [2003] provide a summary of the Athena science payload and Squyres et al. [2006] summarize Opportunity mission results up to embedding into and extrication from the Purgatory ripple, i.e., up to sol 510.

Table 2. Major Activities for Opportunity Organized by Sola
Earth DatebSolActivitySite
  • a

    RS, remote sensing; MEX, Mars Express; AEGIS was an experiment focused on automatic rock detection. Other acronyms defined in Table 1.

  • b

    Read Earth Date 7/1/05–7/4/05 as 1–4 July 2005, etc.

7/1/05–7/4/05511–514Leave “Purgatory” ripple; RS55
7/5/05–7/7/05515–517Drive east toward “Erebus crater”; RS55
7/8/05–7/9/05518–519RS; recharge batteries56
7/10/05–7/14/05520–524Drive east toward “Erebus crater,” first use of combined short segments of blind driving with small slip check segments; RS56
7/16/05–8/5/05525–545RS; Right front steering actuator diagnostic test; Recharge; Continue drive east toward “Erebus crater”56–57
8/6/05–8/23/05546–562Cobble field: IDD “OneScoop,” “Arkansas” cobble, “Perseverance” cobble, “Reiner_Gamma” soil target, “Fruit_Basket” outcrop, and “Lemon_Rind” and “Strawberry” targets58
8/24/05–9/3/05563–573Anomaly and recovery; RS59
9/8/05–9/11/05578–581Continue drive east toward “Erebus crater”; RS59
9/12/05–9/13/05582–583Arrive “Erebus Highway,” Continue drive east toward “Erebus crater”60
9/12/05–9/21/05584–591Continue drive east toward “Erebus crater; RS60
9/23/05–9/26/05592–595Arrive “Erebus crater”; “South Shetland” Feature: Approach and IDD “Deception” target62
9/27/05–9/29/05596–598Warm Reboot Anomaly, Stand Down, and Recover62
9/30/05–10/1/05599–600RS, 360 degree Panorama62
10/2/05–10/5/05601–604Drive northwest around “Erebus crater”; RS62
10/6/05605Backward drive out of ripple to outcrop62
10/7/05–10/10/05606–609Drive westward around “Erebus crater”; RS62
10/11/05–10/12/05610–611Spacecraft reset; Anomaly and recovery
10/14/05–10/31/05613–630Continue drive westward around “Erebus crater”; RS62
11/2/05–11/9/05631–638Arrive “Olympia” outcrop: Approach and IDD “Kalavrita” and “Ziakas” targets64
11/10/05–11/15/05639–644Approach, IDD, and RS “Antistasi” cobble64
11/18/05–11/20/05647–649RS; IDD Composition and Calibration Target64
11/21/05–11/24/05650–653RS; Drive64
11/25/05–11/27/05654–656IDD unstow failure64
12/1/05–12/8/05660–667RS; IDD diagnostics64
12/9/05–12/11/05668–669RS; Atmospheric observations64
12/12/05670Atmospheric RS; Coordinated photometry campaign with MEX64
12/13/05–12/16/05671–674RS; Atmospheric observations, IDD successfully unstowed64
12/17/05–12/20/05675–678IDD “Williams” target64
12/21/05–12/27/05679–685IDD “Ted” target64
12/28/05–1/1/06686–690Continue IDD “Ted” target, IDD “Hunt”64
1/2/06–1/7/06691–696Continue IDD “Ted” target64
1/10/06–1/12/06698–701“Martian Tai Chi”; Atmospheric and targeted RS64
1/13/06–1/15/06702–704RS; unsuccessful IDD unstow64
1/16/06–1/18/06705–706Coordinated observations with MEX64
1/19/06707Successful bump64
1/20/06–1/21/06708–709“Olympia” Outcrop, “Lower Overgaard” feature: IDD “Scotch” target64
1/22/06–1/23/06710–711Coordinated observations with MEX64
1/24/06–1/26/06712–714Continue “Olympia” Outcrop, “Lower Overgaard” feature: IDD “Scotch” target64
1/27/06–1/28/06715–716“Olympia” Outcrop, “Overgaard” feature: IDD “Branchwater” and “Bourbon” targets64
1/31/06–2/4/06719–723“Olympia” Outcrop, “Overgaard” feature: IDD “Don_Giovani,” “Salzburg,” and “Nachtmusik” targets64
2/5/06–2/6/06724–725Bump to “Roosevelt”64
2/8/06–2/12/06726–730“Olympia” Outcrop, “Roosevelt” feature: IDD “Rough Rider” and “Fala” targets64
2/13/06–2/14/06731–733“Olympia” Outcrop, “Bellemont” feature: IDD “Vicos,” “Tara,” “Chaco,” and “Verdun” targets64
2/15/06–2/16/06734–735IDD stall; RS; short IDD diagnostic activity64
2/17/06–2/20/06736–739Runout; Atmospheric Remote Science and Photometry64
2/21/06–3/2/06740–748Drive toward and along “Payson” outcrop; RS64–65
3/3/06–3/4/06749–750“Payson” outcrop: RS64
3/5/06–3/12/06751–758Continue drive toward and along “Payson” outcrop; RS64–65
3/13/06759Recharge; Atmospheric observations65
3/14/06–3/16/06760–762Drive south toward “Victoria crater”; RS65–76
3/17/06763Atmospheric observations65
3/18/06–3/21/06764–767Continue drive south toward “Victoria crater”; RS65–76
3/22/06–3/24/06768–770Odyssey safe mode; Limited downlink capability66
3/25/06–3/26/06771–772Atmospheric observations; RS66
3/27/06–3/30/06773–776Continue drive south toward “Victoria crater”; RS65–76
3/31/06–4/3/06777–779Recharge; RS67
4/4/06–4/14/06780–790Continue drive south toward “Victoria crater”; RS65–76
4/15/06791IDD “Buffalo Springs” outcrop; RS68
4/16/06–4/25/06792–801Continue drive south toward “Victoria crater”; RS65–76
4/26/06–4/27/06802–803Short drive to potential IDD target outcrop69
4/28/06–4/30/06804–806IDD “Brookville” target69
5/1/06–5/9/06807–815Continue drive south toward “Victoria crater”; RS65–76
5/11/06816Atmospheric observations70
5/12/06817Continue drive south toward “Victoria crater”; RS65–76
5/13/06–5/15/06818–820IDD “Cheyenne” outcrop70
5/16/06–5/19/06821–824Continue drive south toward “Victoria crater”; RS65–76
5/20/06–5/22/06825–827IDD “Alamogordo Creek” soil target71
5/23/06–5/27/06828–832Continue drive south toward “Victoria crater”; RS65–76
5/28/06–6/6/06833–842Opportunity embedded in “Jammerbugt” ripple; Extraction71
6/7/06–6/8/06843–844Continue drive south toward “Victoria crater”; RS65–76
6/9/06845Targeted RS72
6/10/06–6/15/06846–851Continue drive south toward “Victoria crater”; RS; Atmospheric observations72
6/16/09852Begin new flight software uplink72
6/25/06–7/7/06860–872Continue drive south toward “Victoria crater”; RS65–76
7/8/06–7/13/06873–878Drive toward “Beagle crater”73–74
7/14/06–7/16/06879–881IDD “Westport” disturbed soil target and “Fort Graham” undisturbed soil target; RS “Dallas” disturbed soil target and “Waco” outcrop74
7/17/06–7/19/06882–884Runout; RS; Drive toward “Beagle Highway”74
7/20/06–7/26/06885–890Approach, scuff, and IDD “Jesse Chisolm” target; IDD “Joseph McCoy” cobble, “Haiwassee” cobble74
7/27/06891Approach and scuff soil target; RS74
7/29/06–8/1/06893–896IDD and RS “Baltra” outcrop pavement74
8/2/06–8/6/06897–901Recharge; Drive to rim of “Beagle”; RS74
8/7/06–8/9/06902–904Spacecraft fault and recovery74
8/15/06–8/16/06910–911IDD “Isabela” and “Marchena” ripple banding targets74
8/17/06–8/23/06912–918Continue drive south toward “Victoria crater”; RS65–76
8/24/06919IDD shoulder azimuth joint stalled; Diagnostic measurements; scuff soil75
8/25/06–8/27/06920–922IDD stall75
8/28/06–9/2/06923–927IDD diagnostics75
9/3/06928Sol 919 scuff: IDD “Powell” and “Powell's Brother” targets75
9/4/06929Drive toward “Emma Dean” crater75
9/11/06–9/16/06936–941Approach, IDD, and RS “Cape Faraday” possible ejecta target75
9/17/06–9/22/06942–947Continue drive south toward “Victoria crater”; RS65–76
9/23/06948Mobility tests76
9/24/06–9/27/06949–952Continue drive south toward “Victoria crater”; RS65–76
9/28/06–8/29/06953–954Targeted RS: “Duck Bay”76
9/30/06–10/4/06955–959Drive toward “Cape Verde” promontory; RS76
10/6/06–10/10/06961–964IDD “Fogo” target76
10/11/06–10/13/06965–967RS; Recharge76
10/14/06–10/15/06968–969Make room in flash for conjunction data76
10/16/06–10/30/06970–984Solar Conjunction; “Cape Verde” Panorama, IDD “Cha” target; Cape Verde Pan76
10/31/06–11/6/06985–991Continue “Cape Verde” Panorama76
11/7/06–11/18/06992–1002Drive to “Cape St. Mary” promontory; RS76
11/19/06–11/20/061003–1004Untargeted RS76
11/21/06–11/22/061005–1006MGS contact attempts; RS “Cape Verde”; Targeted RS76
11/23/06–11/24/061007–1008Targeted RS76
11/25/06–11/28/061009–1012Drive toward “Bottomless Bay”; RS76–77
11/29/061013MRO coordinated observations; RS76
11/30/061014Drive toward “Bottomless Bay”; RS76
12/2/061016Continue drive toward “Bottomless Bay”; RS; IDD autoplace checkout77
12/3/061017Atmospheric RS77
12/4/061018Ground survey; APXS argon density measurement77
12/5/06–12/6/061019–1020RS “Bottomless Bay”; Atmospheric RS77
12/7/06–12/20/061021–1034Drive closer to “Bottomless Bay”; RS; IDD autoplace checkout77
12/22/06–12/23/061035–1036IDD “Rio de Jeneiro” target77
12/24/061037Phoenix demo77, 78
12/25/061038Continue IDD “Rio de Jeneiro” target77
12/26/061039Drive toward “Bay of Toil”; Atmospheric RS77–78
12/27/061040APXS atmosphere77, 78
12/28/06–1/9/071041–1053Drive to, IDD, and RS “Santa Catarina” cobble78
1/10/071054Atmospheric RS78
1/13/071057APXS argon measurement; Atmospheric RS78
1/14/071058Continue drive toward “Bay of Toil”77, 78
1/16/07–1/17/071060–1061“Bay of Toil” long baseline stereo RS78
1/18/071062RS cobbles, image Comet McNaught78
1/19/071063APXS argon measurement; RS78
1/20/07–1/22/071064–1066Drive toward “Cape Desire”; RS; tests to solve visual odometry “picket fence” problem78
1/23/07–1/25/071067–1069Drive toward tip of “Cape Desire”78
1/26/071070IDD and RS78
1/27/071071Continue drive toward tip of “Cape Desire”; RS78
1/29/071072RS magnets and atmosphere78
1/30/07–1/31/071073–1074Long baseline stereo RS “Cabo Corrientes”78
2/1/07–2/3/071075–1077Approach and RS “Cabo Anonimo”78
2/4/07–2/10/071078–1084Drive toward “Cabo Corrientes”; RS79
2/11/07–2/16/071085–1090Drive to position to image “Cape Desire”; Atmospheric RS79
2/17/07–2/23/071091–1097RS “Cape Desire,” “ExtremaDura” outcrop and “Cape of Good Hope” (“Madrid” and “Alava” outcrops)79
2/24/07–2/28/071098–1102Drive to “Cape of Good Hope”; RS79–80
3/1/071103IDD diagnostic79
3/2/07–3/5/071104–1107Continue drive to “Cape of Good Hope”; RS79–80
3/7/07–3/10/071108–1111Drive toward “Valley Without Peril”; RS “Cape St. Vincent”80
3/11/071112RAT grind test80
3/12/07–3/14/071113–1115Continue drive toward “Valley Without Peril”; RS “Cape St. Vincent”80
3/15/07–3/19/071116–1120Race condition fault; Recover; Rest sols
3/20/07–3/26/071121–1127Continue drive toward “Valley Without Peril”; RS “Cape St. Vincent”80
3/27/071128RAT grind diagnostics81
3/28/07–4/1/071129–1133Continue drive toward “Valley Without Peril”; RS “Cape St. Vincent”80
4/2/07–4/4/071134–1136Approach and IDD “Salamanca” and “Sevilla” soil targets (wind streaks); RS81
4/5/07–4/6/071137–1138East and west photometry81
4/7/07–4/9/071139–1141IDD and RS dark streak; RS81
4/10/07–4/11/071142–1143Drive to second location in dark streak; IDD81
4/13/07–4/19/071144–1150Drive to “Alicante” dark streak soil target, IDD, and RS; RS81
4/20/07–4/21/071151–1152Atmospheric RS81
4/22/07–4/25/071153–1156Drive to “Tierra del Fuego”; Long baseline stereo of “Cape St. Vincent”; RS81
4/26/071157Drive to “Granada”81
4/27/07–4/28/071158–1159Atmospheric RS82
4/29/071160RS; D* checkout82
4/30/071161RS “Malaga” and “Granada”82
5/1/07–5/3/071162–1164Drive toward “Cape of Good Hope”; RS82
5/4/071165Atmospheric RS82
5/5/07–5/8/071166–1169RAT touch test on “Viva La Rata”; IDD and RS82
5/9/071170IDD diagnostics; Drive to “Madrid”82
5/10/071171Drive to “Pedriza” cobble82
5/11/071172Soil thermal inertia experiment82
5/12/07–5/13/071173–1174Drive; Atmospheric RS82
5/14/071175Bump to “Cercedilla”; RS82
5/15/07–5/23/071176–1183“Cercedilla” feature: IDD “Penota” target; RS82
5/24/07–6/13/071184–1204Drive toward “Cape Verde”; RS; D* checkout, Visual Target Tracking checkout83, 84, 85
6/14/07–6/22/071205–1213Long baseline stereo RS85
6/23/07–6/29/071214–1219Drive; RS85
6/30/07–8/20/071220–1270Dust storm; Atmospheric dust monitoring; Limited activity to conserve power85
8/21/07–9/3/071271–1284Drive toward rim of “Victoria crater”; Self Portrait; RS85–86
9/4/07–9/8/071285–1288Drive toward ingress point; RS; Diagnostics86
9/9/071289Drive to “Paulo's Perch”; RS86
9/10/071290Atmospheric RS86
9/11/071291Toe dip (drive into and out of “Victoria crater”)86
9/12/07–9/13/071292–1293Drive into Victoria crater; RS86
9/14/07–9/17/071294–1297Odyssey safe mode; RS86
9/18/071298Odyssey safe mode; Drive toward “Alpha Layer”86
9/19/07–9/21/071299–1301Atmospheric RS86
9/22/07–9/25/071302–1305Approach “Alpha Layer”; RS86
9/27/07–10/10/071307–1320IDD “Steno” layer; RS86–87
10/11/071321Drive to second IDD target on “Steno” layer87
10/12/07–10/18/071322–1327“Steno” layer: IDD and RS “Hall” target87
10/19/07–10/20/071328–1329Drive toward “Smith” layer; RS87
10/21/07–10/22/071330–1331Atmospheric RS87
10/23/07–11/18/071332–1358IDD “Smith” rock outcrop; RS “Sharp” sequence of fine layers; Targeted RS; RAT diagnostics87
11/19/07–11/22/071359–1361IDD “Smith2” rock outcrop; RAT diagnostics87
11/23/07–12/10/061362–1379Continue IDD “Smith2” rock outcrop87
12/11/07–12/12/071380–1381Atmospheric RS87
12/13/071382Drive to “Lyell” layer, “Newell” target; RS87
12/14/071383Atmospheric RS88
12/15/07–12/25/071384–1394IDD “Lyell_1” target; RS88
12/26/071395IDD “Lyell_2” target; RS88
12/27/07–1/2/081396–1401IDD “Lyell_3” target; RS88
1/3/081402Drive to “Smith-Lyell” contact; RS88
1/5/081404IDD “Smith Lyell” contact; RS88
1/6/08–1/7/081405–1406IDD “Lyell_4” target; RS88
1/8/08–1/11/081407–1410IDD diagnostics; IDD “Smith_3”88
1/12/08–1/16/081411–1415IDD “Lyell” side of “Smith-Lyell” contact; RS88
1/17/08–1/19/081416–1418Drive to “Buckland” outcrop88
1/20/08–1/21/081419–1420Atmospheric RS88
1/22/08–2/8/081421–1437IDD “Buckland” outcrop; RS88
2/9/08–2/21/081438–1450Drive to “Gilbert” outcrop; RS; Scuff “Gilbert” outcrop; IDD “Lyell-Exeter” target88
2/22/081451APXS argon; RS88
2/23/08–2/25/081452–1454IDD filter and capture magnets; RS88
2/26/08–2/27/081455–1456IDD “Gilbert A”88
2/28/091457“Gilbert” outcrop: IDD “Dorsal” target88–89
2/29/08–3/2/081458–1460DSN transmitter failure; Runout
3/3/08–3/5/081461–1463“Gilbert” outcrop: IDD “Dorsal” target88–89
3/6/08–3/10/081464–1468“Gilbert” location: IDD “Dorsal Tail” target89
3/11/08–3/14/081469–1471“Gilbert” location: IDD “Dorsal New” target89
3/15/08–3/16/081472–1473Supperres rimshot; Atmospheric RS89
3/17/081474Atmospheric RS89
3/18/08–3/26/081475–1483IDD “Gilbert_RAT” target; RS89
3/27/08–4/14/081484–1502Drive toward “Cape Verde”; RS89–90
4/15/08–5/28/081503–1544IDD diagnostics; RS89
5/29/08–5/30/081545–1546RS “Garrels” panorama and “Williams” target; Atmospheric RS89
5/31/08–6/14/081547–1561Continue drive toward “Cape Verde”; RS89–90
6/15/081562Scuff and RS soil90
6/16/08–6/22/081563–1569Continue drive toward “Cape Verde”; RS89–90
6/23/08–7/5/081570–1581“Cape Verde” panorama; RAT diagnostics90
7/6/08–7/24/081582–1600Continue drive toward “Cape Verde”; RS; RAT calibration89–90
7/25/081601Atmospheric RS90
7/26/08–7/28/081602–1604Drive (left front wheel) diagnostics90
7/29/08–7/31/081605–1607RS “Eugene Smith,” “Siever,” and “McKee” targets and “Cape Verde”; Atmospheric RS90
8/1/08–8/3/081608–1610Safe mode90
8/4/08–8/16/081611–1622Exit “Victoria crater”; RS90
8/17/081623RS “Logan” rock weathering target90
8/18/08–8/19/081624–1625Continue exit “Victoria crater”; RS90
8/20/081626RS “Jin” cobble90
8/21/081627Continue exit “Victoria crater”; RS90
8/22/081628RS “Barghorn” target90
8/23/08–8/24/081629–1630Continue exit “Victoria crater”; RS90
8/25/081631RS “Dawson” and “Eugster” targets90
8/26/081632Continue exit “Victoria crater”; RS90
8/28/08–8/29/081634–1635Atmospheric RS90
8/30/081636RS tracks, ripple, and “Isle Royale” target90
8/31/08–9/1/081637–1638Atmospheric RS90
9/2/08–9/3/081639–1640RS “Bright Patch Two” target90
9/4/081641Bump to “Bright Patch”90
9/5/08–9/7/081642–1644IDD “Victoria” rim sand dune90
9/8/08–9/10/081645–1647IDD “Victoria” ripple soil90
9/11/08–9/18/081648–1654Drive toward lee side of ripple; Atmospheric RS90
9/19/08–9/22/081655–1658Atmospheric RS90
9/23/08–9/24/081659–1660RS; APXS Argon90
9/25/081661Drive by “Sputnik crater”90
9/26/08–10/9/081662–1675Drive toward “Cape Victory” and “Cape Agulhas” on “Victoria crater”; RS “Cape Pillar” and “Cape Victory”90–91
10/10/081676RS “Savu Sea” bedrock91
10/11/08–10/13/081677–1679Continue drive toward “Cape Victory” and “Cape Agulhas” on “Victoria crater”; RS “Cape Pillar” and “Cape Victory”90–91
10/14/081680MTES shake91
10/15/081681Continue drive toward “Cape Victory” and “Cape Agulhas” on “Victoria crater”; RS “Cape Pillar” and “Cape Victory”90–91
10/16/08–10/17/081682–1683Drive toward “Endeavour crater”; RS91
10/18/08–11/17/081684–1713Continue drive toward “Endeavour crater”; RS; APXS Argon91–92
11/18/08–11/22/081714–1718Arrive at solar conjunction location; IDD “Santorini” cobble94
11/23/08–11/25/081719–1721“Santorini” Panorama; RS “Corfu” outcrop patch94
11/28/08–12/15/081724–1740Solar Conjunction; IDD “Santorini” cobble94
12/16/08–12/17/081741–1742Delete data products; Atmospheric RS; IDD “Santorini” cobble94
12/18/08–12/22/081743–1747IDD and RS “Santorini” cobble94
12/23/081748Approach “Crete” bedrock and soil targets94
12/24/08–12/25/081749–1750Atmospheric RS94
12/26/08–12/28/081751–1753IDD “Crete” bedrock target94
12/29/081754“Crete” bedrock: IDD “Candia” rock target94
12/30/08–1/2/091755–1758“Crete”: IDD “Minos” soil target94
1/3/09–1/5/091759–1761Atmospheric RS94
1/6/09–1/12/091762–1767RAT diagnostics; Atmospheric RS94
1/13/091768MTES shake test94
1/14/091769RS; APXS Argon94
1/15/091770Drive toward “Ranger crater”94
1/16/091771MTES shake test94
1/17/09–1/20/091772–1775Atmospheric RS94
1/21/091776Approach “Ranger crater”94
1/22/09–1/24/091777–1779RS “Ranger crater”; Atmospheric RS95
1/25/09–1/26/091780–1781Drive south; RS; APXS Argon95
1/27/091782Drive by “Surveyor crater”; RS95
1/28/091783RS; APXS Argon95
1/39/091784RS ripple offset95
1/30/09–2/1/091785–1787Drive; RS; APXS Argon95
2/2/09–2/4/091788–1790PMA fault diagnostics96
2/5/09–2/14/091791–1799Drive toward “Endeavour crater”; RS; APXS Argon97
2/15/091800Right front wheel diagnostic drive98
2/16/091801Automatic AutoNav Map Load Test98
2/17/09–2/23/091802–1808Continue drive toward “Endeavour crater”; RS; APXS Argon98
2/24/091809“Marsquake” observation98
2/25/091810RS; APXS Argon98
2/26/091811FSW R9.3 build98
2/27/09–3/4/091812–1817Continue drive toward “Endeavour crater”; RS; APXS Argon98
3/5/091818RS cobble98
3/6/09–3/9/091819–1822RS; APXS Argon99
3/10/091823Bump to “Resolution crater”; RS99
3/11/091824Approach “Cook_Islands” outcrop99
3/12/09–3/18/091825–1831Approach, IDD, and RS “Cook Islands” cobble patch; RS “Lost” cobble; Atmospheric RS99
3/19/09–4/6/091832–1849“Cook Islands” cobble patch: IDD “Penrhyn” and “Takutea” targets99
4/7/09–4/9/091850–1852Drive to and RS “Adventure” crater99
4/10/091853Atmospheric RS99
4/11/091854Drive to “Discovery crater”99
4/12/091855Atmospheric RS99
4/13/09–4/15/091856–1858Drive to and RS “Pembroke crater”99
4/16/09–5/2/091859–1874Continue drive toward “Endeavour crater”; RS; APXS Argon99–100
5/3/091875“Marsquake” observation100
5/5/09–5/11/091877–1883Bump to pebble patch; IDD and RS “Tilos,” “Kos,” and “Rhodes” pebbles100
5/12/09–5/18/091884–1890Drive to and IDD “Kasos” pebble100
5/19/09–5/26/091891–1898Continue drive toward “Endeavour crater”; RS; APXS Argon100–101
5/27/091899MI sky flats test101
5/28/09–6/10/091900–1912Continue drive toward “Endeavour crater”; RS; APXS Argon102
6/11/09–6/15/091913–1917Atmospheric RS103
6/16/09–6/17/091918–1919IDD “Ios” target103
6/18/09–6/24/091920–1926Drive to and IDD “Tinos” target; RS “Donousa,” “Dryma,” “Naxos,” and “Mykonos” targets103
6/25/09–6/28/091927–1930Continue drive toward “Endeavour crater”; RS; APXS Argon103
6/29/09–7/9/091931–1941IDD “Little Beach” and “Absecon” targets103
7/10/09–7/16/091942–1947Continue drive toward “Endeavour crater”; RS103–104
7/17/091948Marsquake observation; RS104
7/18/091949RS; APXS Argon104
7/19/091950Drive to “Kaiko” and “Nereus” craters; RS104
7/21/09–7/30/091952–1961Drive toward “Block Island” meteorite; IDD and RS104–105
7/31/09–8/10/091962–1972“Block Island” meteorite: IDD and RS “New Shoreham,” “Clayhead Swamp,” “Springhouse Icepond,” “Middle Pond”105
8/11/09–8/12/091973–1974Bump to “Vail Beach” soil pebbles; IDD “Vail Beach”105
8/13/09–8/31/091975–1992Bump to “Siah's Swamp” and “Veteran's Park” targets: IDD and RS; IDD and RS “Siah's Swamp2” and “Fresh Pond”; RS “Block Island” meteorite105
9/5/09–9/11/091997–2003Drive around and image “Block Island” meteorite (6 positions), RS105
9/12/09–9/18/092004–2010Continue drive toward “Endeavour crater”; RS; APXS Argon105
9/19/092011RS “Nautilus” crater106
9/20/09–9/21/092012–2013Continue drive toward “Endeavour crater”; RS106
9/22/09–9/24/092014–2015RS Gjoa crater106
9/25/092016IDD “Limnos” target106
9/26/092017Continue drive toward “Endeavour crater”; RS106
9/27/092018“Marsquake” observation106
9/28/09–9/30/092019–2021Continue drive toward “Endeavour crater”; RS; APXS Argon106–107
10/1/092022RS “Shelter Island” meteorite107
10/2/092023RS; APXS Argon107
10/3/09–10/12/092024–2033Approach, IDD, and RS “Shelter Island” meteorite: “Dering Harbor” target107
10/13/09–10/17/092034–2038Drive to “Mackinac” meteorite; RS107
10/18/09–10/22/092039–2043RS; APXS Argon107
10/23/092044DSN station down, Runout108
10/24/09–10/25/092045–2046Continue drive toward “Endeavour crater”; RS108
10/26/092047RS “Trinidad” crater108
10/27/09–11/3/092048–2054Continue drive toward “Endeavour crater”; RS; APXS Argon108
11/4/09–11/12/092055–2063Drive toward “Marquette Island” rock; RS109
11/14/09–11/20/092065–2071“Marquette Island” rock: IDD “Peck Bay” target109
11/21/09–12/4/092072–2085“Marquette Island” rock: IDD “Islington Bay” target109
12/5/09–12/7/092086–2088Drive toward “Marquette Island's” unseen side109
12/9/09–12/12/092089–2092“Marquette Island” rock: IDD “Loon Lake” target109
12/17/09–1/10/102097–2121“Marquette Island” rock: IDD “Peck Bay” target109
1/11/10–1/13/102122–2124Continue drive toward “Endeavour crater”; RS109
1/15/10–1/19/102125–2129Drive toward “Concepciòn” crater109–111
1/20/102130AEGIS checkout110
1/21/10–1/28/102131–2138Continue drive toward “Concepciòn” crater109–111
1/30/102140“Concepciòn” panorama111
1/31/102141IDD “Mahany Island” target111
2/1/102142“Concepciòn” panorama111
2/2/10–2/3/102143–2144IDD “Loboc River” target111
2/4/10–2/7/102145–2148Drive toward “Concepciòn crater”111
2/8/102149Bump toward “Chocolate Hills”111
2/9/10–2/10/102150–2151“Chocolate Hills” rock: IDD “Aloya” dark material111
2/11/102152Bump around “Chocolate Hills”111
2/12/10–2/14/102153–2155“Chocolate Hills” rock: IDD “Arogo” target111
2/16/10–2/19/102157–2160“Chocolate Hills” rock: IDD “Tears” target, IDD “Dano” target111
2/20/10–3/3/102161–2171Drive around “Concepciòn crater”; RS111
3/4/102172AEGIS checkout111
3/5/10–3/8/102173–2176Continue drive around “Concepciòn crater”; RS111
3/9/102177Return to “Pink Path” and drive to “Endeavour crater”111
3/10/10–3/27/102178–2195Continue drive toward “Endeavour crater”; RS111–112
3/28/10–3/30/102196–2198RS “San Antonio West” and “San Antonio East” craters114
4/1/10–4/12/102199–2210Continue drive toward “Endeavour crater”; RS114
4/13/102211Dune crossing, soil mechanics experiment115
4/14/10–4/22/102212–2220Continue drive toward “Endeavour crater”; RS115
4/23/102221AEGIS watch115
4/26/102224IDD “Ocean Watch” undisturbed soil target116
4/27/10–4/29/102225–2227Continue drive toward “Endeavour crater”; RS116
4/30/10–5/2/102228–2230Drive toward “Newfoundland” rock; RS116
5/3/10–5/13/102231–2240Continue drive toward “Endeavour crater,” drive to “Lily Pad”; Argon APXS, drive to “Lily Pad”; RS116
5/14/10–5/19/102241–2246RS; MarsQuake Experiment; Drive south116–117
5/20/10–5/26/102247–2253AEGIS experiment; Drive east; RS117
5/27/10–5/29/102254–2256Drive; RS117
5/30/10–5/31/102257–2258Get fine attitude; recharge; Mini-TES in stand down mode117–118
6/1/10–6/3/102259–2261PMA diagnostics118
6/4/10–6/6/102262–2264Hazcam; Argon APXS; Recharge118
6/7/102265PMA diagnostics118
6/8/10–6/9/102266–2267RS; Recovery QFA and postdrive imaging118
6/10/10–6/17/102268–2274RS; Recharge; Drive east toward Endeavour118
6/18/10–6/19/102275–2276RS; Drive south118
6/20/10–6/21/102277–2278Argon APXS; AEGIS experiment119
6/22/10–6/30/102279–2287Drive east-southeast, drive northward, drive east; RS120
7/1/10–7/5/102288–2292Drive eastward; RS; AEGIS experiment120
7/6/10–7/7/102293–2294Drive east toward “Endeavour crater”; RS120
7/8/10–7/9/102295–2296Drive; RS120
7/10/102297IDD “Juneau_Road_Cut” target, IDD “Juneau” target120
7/12/10–7/13/102299–2300Drive east-southeast toward gravel pile; RS120–121

2. Background Discussion

[4] MER Mission science objectives are focused on remote sensing and in situ observations along traverses to characterize current and past Martian environments and the role of water in formation and alteration of the surface and associated crustal materials [Squyres et al., 2003]. These objectives are aligned with the overarching NASA Mars Exploration Program themes of “follow the water” and searching for evidence of past or present habitable zones and life. For reference, on the other side of the planet the Spirit rover has been exploring the Inner Basin, Columbia Hills, Gusev crater, and has acquired data that indicate the presence of hydrated sulfate, opaline, and carbonate-bearing mineral deposits of likely fumarolic or hydrothermal origins [Arvidson et al., 2008, 2010; Squyres et al., 2008; Morris et al., 2010]. Opportunity landed on the Meridiani plains, selected primarily because the Mars Global Surveyor Thermal Emission Spectrometer (TES) data indicated a high abundance of hematite, a mineral typically formed in an aqueous environment [Christensen et al., 2001]. Data collected by Opportunity within Eagle and Endurance craters conclusively showed that the hematite signature is carried by hematitic concretions weathered from sulfate-rich bedrock and concentrated as a surface lag, partly worked into basaltic sand ripples that cover much of the plains [Soderblom et al., 2004; Sullivan et al., 2005; Arvidson et al., 2006; Jerolmack et al., 2006]. Further, the sulfate-rich sandstones that comprise the Burns formation and that underlie the ripples were found to be largely ancient aeolian sandstone deposits, with local reworking within ephemeral lakes [Squyres et al., 2004; Grotzinger et al., 2005, 2006; Metz et al., 2009]. Post depositional aqueous processes have altered the deposits as evidenced by the presence of hematitic concretions and fracture-filling deposits [McLennan et al., 2005; Knoll et al., 2008]. Opportunity has continued to search for the mud-rich rock facies that would help confirm the hypothesis that the sands formed from precursor evaporates in a playa lake environment.

[5] The Burns formation rocks examined by Opportunity are part of a regional-scale deposit that covers several hundred thousand square kilometers and is best explained by accumulation during one or more periods of rising groundwater [e.g., Andrews-Hanna et al., 2010] (Figure 1). These deposits are draped unconformably onto dissected cratered terrain and exhibit an impact crater size frequency distribution indicative of a preservation age of Noachian or Early Hesperian [Arvidson et al., 2006]. Preexisting craters are also evident and show partial burial by the sedimentary deposits, including the ∼20 km diameter Endeavour crater toward which Opportunity is traversing (Figure 1). Bopolu is a ∼19 km diameter crater located to the southwest of Opportunity (Figure 1). This crater clearly postdates the deposition of the sulfate-rich sedimentary deposits, given that the floor and rim of the crater, along with its ejecta deposits, exhibit basaltic signatures [Christensen et al., 2001] and the ejecta deposits extend over the sedimentary deposits (Figure 1). Bopolu, and other rayed craters on Meridiani Planum, lack hematite signatures on their ejecta deposits, implying that these craters formed after the hematite was concentrated on the surface [Golombek et al., 2010]

[6] Detailed measurements conducted by Opportunity in Eagle and Endurance craters showed the utility of impact craters for assessing the stratigraphy of the layered sedimentary rocks within Merdiani Planum [e.g., Squyres et al., 2006]. This approach was continued during the period of the mission covered by this paper, including measurements of strata within Erebus and Victoria craters, together with remote sensing and in situ observations of smaller craters and ejecta deposits encountered during traverses. Opportunity has also continued to characterize aeolian deposits, cobbles, and the atmosphere. Key results for all of these measurements are presented in this paper, along with brief summaries of instrument and vehicle status, and a chronicle of traverses and measurements from sols 511 to 2300.

3. Mission Overview

3.1. Rover and Payload Status

[7] From sol 1 to sol 2300 Opportunity traversed 21,760 m and from sol 511 to sol 2300 16,389 m were covered, based on tracking wheel turns (Figure 2). For reference, Table 1 provides a summary of the Athena science payload, with selected comments about status. Opportunity and its Athena payload were not designed and built to travel over thousands of meters and operate over six and one half years. Even so, the vehicle and payload have continued to operate well.

[8] Opportunity's right front steering actuator failed on sol 433, leaving the wheel rotated inward by an ∼8 degree angle. On sol 654 the Instrument Deployment Device (IDD) experienced an unstow anomaly because of a failing shoulder joint actuator. This joint was declared “failed” on sol 1542 and since then the IDD has been left deployed forward, carried in a “fishing stow” position while driving. IDD deployments have still been possible, but within a more limited work volume as compared to earlier measurements. Wheel currents for the right front wheel have occasionally spiked during drives, perhaps because of uneven distribution of lubricant. The solution has been to rotate the affected wheel backward and forward to even out the lubricant and to “rest” the wheel when currents spiked to particularly high values. The vehicle has primarily been driven backward during the mission period covered by this paper, in part to minimize wheel actuator current spikes, and because this mode was found to permit Opportunity to cross ripples with minimal mobility difficulties.

[9] Opportunity has survived four Martian summers, with their associated dust storms and periods of high atmospheric opacity (Figure 3). Solar array energy has varied widely from low values during the winter to high values during the summer, with strong modulations based on the amount of dust in the atmosphere and on the solar panels. Dust accumulation on the panels was predicted to end rover life much earlier, but winds have removed dust on sols 520, 1150, 1305, 1520 (minor), 1620, 1846 (minor), 1990 (minor), and 2300, providing instantaneous increases in available energy. Unlike Spirit, which is located at −14.57° latitude, Opportunity's near equatorial location (−1.95° latitude) has provided enough sunlight to allow the vehicle to continue operations throughout the winter seasons, although at a reduced pace relative to summer operations.

Figure 3.

Time series available rover energy, atmospheric opacity, and major events for Opportunity from sol 511 to sol 2300. Opacity at 880 nm is shown multiplied by a factor of 200.

[10] Dust coatings on the Pancam optical surfaces have made radiometric calibration of images a continuing and involved process, particularly because the Pancam calibration target on the rover deck has also accumulated dust. Dust has accumulated on the Navcam and Hazcam exterior optical surfaces, but has not compromised the use of these cameras for either scientific purposes or operations, including driving and IDD deployment planning. During calendar year 2007 (∼sol 1240, Figure 3) a global dust storm deposited dust on the Microscopic Imager (MI) exterior optics, both inside and outside the protective dust cover. The dust cover seal is not airtight as it was designed to allow gas to escape during launch. Consequently, useful MI images can no longer be acquired with the dust cover closed, and images taken with the dust cover open are visibly affected by dust contamination. This contamination has reduced the signal/noise in Opportunity MI images, but has not affected the ability to retrieve useful textural information from the data. Unfortunately, dust accumulation on the Mini-TES exterior mirror compromised the ability to retrieve quantitative information about mineralogy from data acquired after ∼sol 1217. The instrument ceased responding to commands from the rover on sol 2257.

[11] The Alpha Particle X-Ray Spectrometer (APXS) has continued to operate nominally, acquiring compositional information for soils and rocks and making measurements of atmospheric argon. The Mössbauer Spectrometer (MB) has also continued to acquire data for soils and rocks, although significant decay of the cobalt-57 radioactive source (271.79 day half-life) eventually required measurements extending over several sols to achieve appropriately high spectral signal to noise values. The Rock Abrasion Tool (RAT) has ground into 38 rocks over the course of the mission. During the period covered by this paper 18 rocks were brushed and 15 were ground. By sol 2300 the grinding bit pads were worn to ∼20 to 30% of their original thicknesses and the brush was slightly bent, no longer sweeping out a complete circle. All three RAT encoder motors have stopped operating, leading to step-by-step manual approaches for commanding the RAT to avoid a brush or grind failure or damage to the instrument.

3.2. Overview of Mission Activities

[12] Table 2 provides a sol-by-sol description of Opportunity's activities. Figure 3 shows a timeline of major activities with available solar panel energy and atmospheric opacity on a sol-by-sol basis. Traverses have been mainly from north to south, with stops in or near craters to examine rock strata and other “jogs” to examine important science targets or avoid large ripples (informally termed “purgatoids” after the Purgatory ripple, where Opportunity was embedded between sols 446 and 484) (Figure 2). By ∼sol 2250, after bypassing several fields of purgatoids, Opportunity was able to start traversing southeast, directly toward the rim of Noachian-aged Endeavour crater, with the intent of exploring the ancient rocks exposed on the crater's rim.

[13] The initial primary science target for Opportunity, after leaving Purgatory, was Erebus, a ∼220 m wide, degraded impact crater with ripples surrounding the crater and occupying most of the crater floor (Figure 2). Extensive remote sensing and in situ measurements were made on the Olympia outcrop on the northwestern side of Erebus crater. Detailed remote sensing data were also acquired for the Payson outcrops located on the southwestern side of Erebus. The Jammerbugt ripple was another location in which significant wheel sinkage and slippage were encountered. Opportunity backed out of this ripple on sol 841, after six sols of extrication. Several rock and soil targets, along with small craters, were then examined during traverses to Victoria crater. Opportunity reached Victoria's crater rim on sol 953.

[14] The first part of the Victoria campaign focused on remote sensing of crater walls, done by driving to promontories and acquiring Pancam and Navcam mosaics of adjacent cliff faces, together with acquiring data for cobbles strewn onto the smooth annulus surrounding the crater (Figure 2). This phase focused on the northwestern quadrant of the crater and extended from sols 953 to 1290. After determining that Duck Bay was a reasonable location to enter the crater and make measurements of Burns formation rock targets as a function of depth, Opportunity entered Victoria on sol 1292 and stayed in the crater, acquiring detailed remote sensing and in situ observations. Opportunity exited Duck Bay on sol 1622. After soil measurements on the annulus surrounding Victoria, and additional remote sensing of the crater walls, Opportunity on sol 1682 began traverses toward Endeavour crater, located ∼20 km to the southwest of Victoria.

[15] During the post-Victoria traverses Opportunity encountered a number of impact craters and a variety of cobbles and boulders, discussed in detail in a subsequent portion of this paper. Key to successful traversing has been the avoidance of purgatoids, using Mars Reconnaissance Orbiter HiRISE images [McEwen et al., 2007] and Odyssey THEMIS-based thermal inertia maps [Christensen et al., 2004a; Fergason et al., 2006], together with extensive rover-based remote sensing observations, to drive along paths that minimized traverses across large ripples [Parker et al., 2010]. This led to a southern to southwestern path south from Victoria to avoid the purgatoid fields, and then turning to the southeast on a more direct route to the rim of Endeavour (Figure 2).

4. Modern Atmospheric Processes

[16] In addition to addressing the primary mission themes of characterizing past environmental conditions, and the role of water in formation and alteration of crustal materials, Opportunity has continued to make periodic measurements that pertain to current atmospheric characteristics and dynamics. Pancam has been used on a sol-by-sol basis to determine atmospheric opacity at wavelengths of 440 and 880 nm. This activity has supported mission operations tactically by providing estimates of available solar irradiance on the panels. When combined with orbital observations, the Pancam measurements have been critically important for tracking dust storms and the impact on vehicle performance and safety. Regional dust storms have been observed during each summer season, with a planet-encircling event occurring during the early summer of the third year of operations (∼sols 1200–1300, see Figure 3). This led to an available energy of less than 200 Wh and placed Opportunity in a survival mode. Occasional dust devils have been imaged by Opportunity and repeated coverage of the plains surrounding Opportunity by the Mars Reconnaissance Orbiter's Context Imager (CTX) [Malin et al., 2007] showed a number of ephemeral dust devil tracks. Wind directions and magnitudes were found by inspection of aeolian streaks, and mesoscale modeling of atmospheric circulation models, to vary during the course of the Martian year [Sullivan et al., 2005; Jerolmack et al., 2006; Geissler et al., 2010; M. Chojnacki et al., Orbital observations of contemporary dune activity in Endeavour crater, Meridiani Planum, Mars, submitted to Journal of Geophysical Research, 2011].

[17] Winter atmospheric measurements for science have focused on sky imaging to detect the well-known winter aphelion water ice cloud “belt” [e.g., Clancy et al., 1996]. Fewer clouds were observed during the three periods near aphelion than would have been predicted from Earth-based and orbital observations [i.e., Wolff et al., 1999]. The paucity of clouds suggests that the meteorology of the Meridiani region may be more complex (and thus more interesting) than previously understood.

[18] When not in use to measure compositions of rocks and soils, and when rover energy permitted, APXS has been used to monitor seasonal and interannual variations in atmospheric argon contents (Figure 4). The argon mixing ratio is a tracer for atmospheric transport because it is a noncondensable gas under Martian conditions. On the other hand, the carbon dioxide content of the atmosphere varies significantly as a function of season because it condenses over the winter pole to form the seasonal ice cap. The southern winter is longer and colder than the northern winter season because the southern winter season occurs near aphelion. The southern pole is also topographically higher than the northern pole. Consequently, the southern winter carbon dioxide cap is more extensive than the northern winter cap. These dynamics are evident in global pressure variations recorded by the Viking Landers, with lowest surface pressures associated with the southern winter [Tillman et al., 1993].

Figure 4.

Scaled volume mixing ratios of argon to carbon dioxide are shown for three years of Opportunity APXS observations of the atmosphere together with mixing ratios derived from the NASA Ames general circulation model. Both data sets have been scaled by subtracting means and dividing by standard deviations. The southern winter solstice occurs at Ls = 90°, whereas the northern winter solstice occurs at Ls = 270°. See text for explanation of trends.

[19] Sprague et al. [2007] reported global variations in argon mixing ratios using Odyssey gamma ray spectrometer data, focusing on the sixfold argon enhancement during the winter over the southern polar region (−75 to 90° latitude). Argon mixing ratios were found to peak over the south pole at Ls = 90 degrees and then undergo a rapid decline to lowest values by the southern summer (Ls = 270 degrees). Argon mixing ratios for an adjacent latitude band (−60 to 75 degrees latitude) showed a similar pattern, but shifted to a slightly later time. Argon was not found to concentrate above the north polar winter cap. These patterns were interpreted as evidence for meridional transport of argon with carbon dioxide as part of the global atmospheric circulation system, particularly combined with relatively weak south pole to equator transient eddies that cause buildup of argon over the winter cap [Nelli et al., 2007].

[20] Opportunity-based argon atmospheric mixing ratios show minimum values at Ls = 90 degrees (beginning of the southern winter), during the period when argon shows the highest concentration over the growing south polar seasonal cap (Figure 4). The Opportunity-based argon mixing ratios increase rapidly as the south polar concentrations decrease. Peak Opportunity-based values are reached during late southern winter to early spring seasons. Argon mixing ratios then start decreasing, reaching a broad low between 270 to 320 degrees Ls (northern winter season). The overall trends were simulated with the NASA Ames global circulation model, which reproduced the broad patterns discussed above, including the sharp decrease and increase in Opportunity-based values associated with the south polar winter cap formation and sublimation. The broad low associated with formation and sublimation of the northern winter cap is also reproduced. The model fit to the data shows temporal offsets and there are also interannual variations in Opportunity-based argon mixing ratios. The differences between the model and the interannual variations are currently study topics. It is clear that the Opportunity data provide quantitative “ground truth” for use with Odyssey-based argon mixing ratios as tracers for atmospheric circulation, along with “ground truth” validation of global circulation models.

5. Aeolian Ripples and Mobility Issues

5.1. Nature and Origin

[21] The plains surfaces traversed by Opportunity are largely covered by aeolian ripples that have crests predominantly oriented in a north to south direction. Hematitic concretions are concentrated on the ripple crests, whereas interiors are dominated by a mix of basaltic sand, hematitic concretions, and dust [Christensen et al., 2004b; Soderblom et al., 2004; Sullivan et al., 2005; Arvidson et al., 2006; Jerolmack et al., 2006]. Opportunity's remote sensing measurements during traverses and stops for in situ measurements during the period covered by this paper show a change to larger ripples and a greater areal exposure of outcrops, beginning ∼280 m south of Purgatory ripple (Figure 2). This change corresponds to a boundary between a relatively uniformly dark, high thermal inertia surface covered by relatively small ripples to the north, to a mix of bright and dark surfaces to the south (Figure 2). The darker surfaces south of the boundary, which are dominated by ripples, have low thermal inertias relative to the surrounding brighter surfaces, which are dominated by bedrock exposures. South of Victoria fields of large ripples (i.e., purgatoids) identified using HiRISE images were avoided during traverses by initially traversing to the southwest, then south, and finally back to the southeast [Parker et al., 2010].

[22] Detailed in situ measurements of ripple surfaces and interiors collected throughout the Opportunity mission provide key information on the nature and extent of these aeolian features. For example, on sol 2297, at the end of a drive, Opportunity turned to maximize UHF communication rates. The turn caused the right front wheel to cut into the crest of a ripple, exposing the upper part of the interior (Figure 5). MI data were acquired for the excavated materials and an undisturbed surface target, Juneau, on the western flank of the ripple. The Juneau images show a high areal concentration of hematitic concretions (Figure 6), consistent with prior observations of ripple surfaces (Figures 7). Specifically, application of correspondence analysis [e.g., Arvidson et al., 2006, 2008, 2010] and comparison to MI data show that the compositional trends associated with factor 1 loadings range from basaltic sand samples (e.g., Auk target within Endurance crater) as one end-member, to almost complete areal coverage by hematitic concretions (e.g., Juneau) as the other end-member. In fact, the location of samples along this mixing line (i.e., factor 1 loading in Figure 7) is predicted with high fidelity based on the areal fraction of concretion coverage. A minor, but important direction (factor 2 loading shown in Figure 7) delineates targets with enhanced sulfur and chlorine. MI data for these samples show the presence of fine-grained materials (e.g., Les Houches, Figure 6). These fine-grained targets are found adjacent to Eagle (sol 60 data), Endurance (sol 123), and Victoria (sol 1647) craters and include dust likely deposited in or near the craters in local aeolian traps.

Figure 5.

Front Hazcam image looking to the north showing a drive and turn in place across the crest of a ripple. Outcrop is shown as bright polygonal regions on either side of the ripple. Juneau is the location of an Instrument Deployment Device (IDD) target on the western flank of the ripple. Microscopic Imager (MI) data for Juneau show that it is covered with a dense array of hematitic concretions (Figure 6). Front Hazcam frame 1F331928037RSLAKVTP1214L0MZ acquired on sol 2295.

Figure 6.

Series of MI images, each 3 cm across, of soils encountered during Opportunity's traverses. Juneau is a ripple surface dominated by hematitic concretions. Auk sand was encountered in Endurance crater, whereas the fine-grained dust target, Les Houches, was found on the perimeter of Eagle crater. These targets form compositional end-members in the APXS data, as shown in Figure 7.

Figure 7.

Correspondence analysis plot showing all undisturbed soil targets (labeled by sol) for which APXS data were acquired. Numbers in parentheses correspond to the ratio of ferric to total iron for the targets based on MB observations. The primary trend is from basaltic sands on the right side of the diagram to high concentrations of hematitic concretions on the left. Targets with enrichments in dusty material map separately from the dominant sand to concretion “mixing line.” Iron oxidation states are consistent with the inferred mineralogy, with higher values for targets with a higher abundance of hematitic concretions.

[23] Ripples within the regions traversed by Opportunity have a dominant strike of north-south, based on examination of HiRISE data and azimuths measured from Pancam and Navcam data (Figures 810). The Raleigh crater (Figure 9) is part of the Resolution crater cluster, a group of small impact craters spread over an area of ∼120 m by 80 m, and is located to the south of Victoria (Table 2). Raleigh (and others within the cluster) must have formed after the last major phase ripple migration ceased, since the crater cuts across the ripples and exposes layers perpendicular to the ripple crest (Figure 9) [Golombek et al., 2010]. These exposures provide important information about the wind direction that produced the ripple fields. The layers dip slightly toward the west. This pattern is consistent with ripple formation by easterly winds in which sand was trapped on the leeward faces and the ripple migrated over the deposits, producing layers that dip slightly toward the leeward direction. The concentration of hematitic concretions on crests is due to the fact that these relatively large grains travel in creep or traction mode and are left behind as the sand saltates in the wind direction. Currently the dominant sediment-moving winds vary in azimuth over time [e.g., Geissler et al., 2010; M. Chojnacki et al., submitted manuscript, 2011] and seem to have produced a set of subsidiary ripples and serrated the older, larger ripple crests, depositing fine-grained materials within and on the large north-south oriented ripples (e.g., Figure 10). This trapping of fine-grained material is inferred to have produced surfaces with slightly lower thermal inertias as compared to surrounding bedrock or smaller ripples.

Figure 8.

HiRISE view of ripples around Raleigh crater, along with Opportunity's traverses and sols shown for selected positions. Note the bright outcrop and extensive coverage by north-south trending aeolian ripples. Portion of HiRISE frame ESP_016644_1780_red.jp2.

Figure 9.

Navcam view looking to the south into the ∼2 m wide Raleigh crater. Note the eastern ripple slope exposes light and dark bands that can be extended to the third dimension by noting the cross stratification on the crater wall. Stratification is consistent with formation of ripples by easterly to southeasterly winds, as discussed in the text. Image acquired on sol 1852. Navcam frame 1N292594992RSD99NGP1921L0MZ.

Figure 10.

Pancam false color image mosaic looking north and showing the bright red western sides of ripples. Pancam bands L2 (753 nm), L5 (535 nm), and L7 (432 nm) are shown as red, green, and blue. Data acquired on sol 1858, after a major dust storm.

[24] On a regional scale in Meridiani Planum low thermal inertia streaks are common and extend to the west and northwest from craters and other topographic obstacles, including the locations of the large ripples to the west and northwest of Endeavour crater (Figure 11). These low-temperature streaks have remained invariant during the 2002 to 2010 observation period of the Odyssey THEMIS IR observations and are interpreted to have been produced by easterly to southeasterly winds. Global circulation models that simulate the modern climate of Mars do not produce strong easterlies within the latitudes that include Meridiani Planum [Fenton and Richardson, 2001; Haberle et al., 2003]. Indeed, observations of fresh impact craters seen by Opportunity and HiRISE indicate that the latest major phase of ripple migration occurred between ∼50,000 to 200,000 years [Golombek et al., 2010]. During earlier periods of time, when the spin axis of Mars was at a higher orbital obliquity than the current value of 25 degrees, the solstice Hadley cell circulation would have occupied a wider latitudinal belt and very likely produced strong easterlies that generated the north-south oriented ripples observed by Opportunity. In fact, modeling of obliquity changes shows that the obliquity can vary by 20° over a time scale of hundreds of thousands of years [e.g., Ward and Rudy, 1991], a timing consistent with the last major ripple migration period as discussed above.

Figure 11.

Color-coded predawn THEMIS IR scaled temperature values and Opportunity's traverses overlain onto THEMIS daytime IR mosaic. Arrows show low-temperature streaks extending eastward from craters. Note also the low-temperature zone to the west of Endeavour, including regions traversed by Opportunity. Red colors correspond to terrains with thermal inertias between ∼155 and 180 J m−2 K−1 s−1/2 and blue to regions with thermal inertias between ∼140 and 145 J m−2 K−1 s−1/2. Box shows region covered in Figure 2. Lower thermal inertia areas are dominated by fields of relatively large ripples that have trapped fine-grained aeolian deposits. Bedrock has slightly higher thermal inertias than ripples.

5.2. Mobility Issues

[25] Crossing ripples with wavelengths larger than Opportunity's wheel base has occasionally proven to be problematic, leading to embedding at the Purgatory and Jammerbugt ripples, and causing excessive wheel sinkage and slippage as recently as sol 2220 (Table 2 and Figures 1213). These mobility difficulties occurred while Opportunity attempted to drive up and over ripple flanks. To provide a quantitative evaluation of these mobility difficulties a 200 element dynamical model of Opportunity was constructed in software, including wheel-soil interactions with wheel sinkage and slippage into deformable soils. This software was built on the framework developed for modeling Spirit and its proposed extrication drives [Arvidson et al., 2010]. Normal and shear stresses between the wheels and soil were modeled using the classical Bekker-Wong terramechanics expressions that describe relationships among normal and shear stresses, applied wheel torque, wheel slip, and wheel sinkage as a function of soil properties [e.g., Wong, 2003]:

equation image
equation image

where σ is the normal stress and τ is the shear stress between the wheel and soil, kc/b is the ratio of soil cohesion moduli to wheel width, kϕ is the internal friction moduli, z is the depth of wheel sinkage, n is a scaling exponent, c is the soil cohesion, ϕ is the soil angle of internal friction, j is the slip value between the wheel and soil, and kx is the shear deformation modulus in the longitudinal or drive direction. The value for j is determined based on the magnitude of wheel sinkage into soil.

Figure 12.

HiRISE view of sol 2220 high slippage and sinkage location on the western side of a ripple. Locations A and B are shown on the Navcam view in Figure 13. Portion of HiRISE frame ESP_016644_1780_red.jp2.

Figure 13.

Navcam image acquired on sol 2226 of the sol 2220 high slippage and sinkage location on the western side of a ripple. The shallow angle of attack relative to the ripple crest put all six wheels on the relatively soft soil. During its climb up the ripple, slip equaled 58% and the drive was halted by onboard Visodom software. Navcam frame 1N325805274RSDAG12P19170L0MZ.

[26] Increased wheel sinkage due to increased weight over a given wheel generally leads to increased contact area between the wheel and soil and increased compaction resistance, thereby increasing the amount of slippage, S, for a driven wheel as motor torques are increased to compensate:

equation image

where V is the longitudinal velocity of the wheel, R is the wheel radius, and ω is the wheel angular velocity. As slippage increases, additional sinkage generally occurs as soil is moved in the direction of the spinning wheel. This further increases motion resistance as the wheel comes in contact with additional soil during sinkage. At some point the maximum soil shear stress before failure is reached and slippage becomes effectively 100%, causing longitudinal motion to cease.

[27] The sol 2220 drive ended when visual odometry [Maimone et al., 2007] showed ∼58% wheel slip, which was above the limit set for continuing the drive. This was a fortuitous event for Opportunity since onboard use of the imaging systems to track slip was done every ∼20 m or so during a traverse. The high slip occurred when Opportunity was driving backward and scaling the western side of the ripple at an acute angle (∼25°) relative to the ripple crest azimuth, with a rover tilt magnitude of ∼8° (Figures 1213). The ∼58% slippage values occurred when six wheels were on the ripple. Wheel sinkage measured from Navcam data taken after extrication from the ripple was ∼5 cm. The front wheel (i.e., on the downslope side) tracks show evidence of slip sinkage, based on disruption of the cleat imprints (Figure 13).

[28] To replicate the incipient embedding the Opportunity dynamic element model was set to drive backward on an 8° degree slope into nearly cohesionless soil (c∼1 kPa) with an angle of internal friction, ϕ = 30°, following results from soil trenching experiments conducted by Opportunity earlier in the mission [Sullivan et al., 2010]. Other parameters in equations (1) and (2) were varied to match actual drive results, although detailed sensitivity calculations showed that results were to first-order invariant to the chosen values for kc. Thus, the modeling focused on varying numerical values of kϕ, n, and kx. The first two parameters control the amount of static sinkage whereas the third controls the amount of slippage for a given sinkage magnitude. The models replicated observed values of sinkage and slippage with n = 1.1, kϕ = 75,000, and kx = 65 mm. These values are consistent with the presence of relatively soft soil into which the wheels for the 179 kg Opportunity rover would sink to a few centimeters on flat terrain. Also the value for kx is toward the upper end of sandy soils for Earth and indicates that relatively high slip values should occur with even modest sinkage and increased compaction resistance. The lesson for mobility was to keep all six wheels from simultaneously being on a ripple flank with the vehicle driving in an uphill direction. During the simulation it was found that the middle and rear wheels bore most of the weight and thus underwent the most sinkage, thereby significantly increasing compaction resistance. Slippage increased as torque was increased to maintain constant wheel angular velocity, leading to slip sinkage, and exceeding the 58% slippage limit for continuing the drive. Opportunity was able to back out of the ripple with one drive and was then commanded to continue to drive south along an interripple zone until a smaller ripple system was encountered to cross over toward the east.

6. Cobble and Boulder-Sized Rock Fragments

[29] During the mission period covered by this paper Opportunity has characterized a number of individual rock fragments with a variety of sizes. These rocks have been found near craters, in isolated clusters covering small to moderate (tens to hundreds m2) areas, and sometimes as isolated pebbles, cobbles, or boulders separated by hundreds of meters. For reference, Table 3 lists all rock fragments that were investigated in detail with IDD instruments throughout the mission. Many additional cobble-sized and smaller rock fragments have been characterized with Pancam, Navcam, and/or Hazcam observations [e.g., Weitz et al., 2010]. These observations are not listed in Table 3.

Table 3. Overview of Cobbles Discussed in This Paper in Order of Discovery
NameaSolbClassification and ReferencecLocationMeasurementsd
Bounce Rock63basaltic shergottite [1, 2]close to the rim of Eagle crater2U, R7U, R10U,4R
Lion Stone105outcrop fragment [2, 3]northwestern rim of Endurance craterU, RU, R2U, 5R
Barberton121dark toned cobble (Barberton group) [2, 3, 4, 5]southern rim of Endurance craterUU2U
Heat Shield Rock347iron meteorite [2, 4, 6, 7]plains; close to the Heat ShieldU, BU, B6U, 4B
Russett381outcrop fragment [2, 3]plainsUU2U
Arkansas551dark toned cobble (Arkansas group) [2, 3]close to Erebus craterUUU
Perseverance554dark toned cobble (Arkansas group) [2, 3]close to Erebus craterUn. a.U
Antistasi641dark toned cobble (Arkansas group) [2, 3]close to Erebus craterUU4U
JosephMcCoy886dark toned cobble (Arkansas group) [2, 3]Jesse Chisholm area close to Beagle craterUU4U
Haiwassee886dark toned cobble (Arkansas group) [2, 3]Jesse Chisholm area close to Beagle craterUn. a.U
Santa Catarina1045dark toned cobble (Barberton group) [2, 3, 4, 5]Cobble field close to the rim of Victoria craterUU5U
Santa Catarina cobble field1045likely genetically related to Santa Catarina [4, 8]Cobble field close to the rim of Victoria cratern. a.n. a.n. a.
Santorini1741dark toned cobble (Barberton group) [2, 3, 5]∼800 m south of Victoria crater2UU3U
Kos1879dark toned cobble (Arkansas group) [2]∼1.5 km south of Victoria craterUn. a.U
Tilos1879dark toned cobble (Arkansas group) [2]∼1.5 km south of Victoria craterUn. a.U
Rhodes1879dark toned cobble (Arkansas group) [2]∼1.5 km south of Victoria craterUn. a.U
Kasos1886dark toned cobble (Barberton group) [2, 3, 5]∼1.5 km south of Victoria craterUU4U
Block Island1961Iron meteorite [2, 4, 7]∼4 km south-southwest of Victoria crater6U4U26U
Vail Beach1974dark toned cobble (Arkansas group) [2, 3]∼4 km south-southwest of Victoria craterUn. a.5U
Shelter Island2022Iron meteorite [4, 7] Un. a.2U
Mackinac Island2034Iron meteorite [4, 7] n. a.n. a.n. a.
Marquette Island2065Martian mafic igneous ejecta block [9] 3U, R3U, R5U, 8B, 10R
Chocolate Hills2150Impact melt covered ejecta blockon the rim of Concepción craterUU18U

[30] Five basic types of rock fragments were found and characterized in detail during Opportunity operations: (1) local impact ejecta that consist of sulfate-rich sedimentary material (e.g., Chocolate hills, Figures 1415); (2) basaltic materials that are likely impact ejecta fragments (e.g., Bounce Rock) from distant sources; (3) rock fragments that are a mix of sulfate and basaltic materials that are likely impact melt products (e.g., Arkansas); (4) stony-iron meteorites (e.g., Barberton); and (5) iron-nickel meteorites (e.g., Block Island, Figure 16).

Figure 14.

Pancam false color view of Concepción crater acquired on sol 2140, including the Chocolate Hills and Loboc River boulders, when Opportunity was ∼5 m from the northern rim. Ejecta from this impact event are superimposed on the surrounding ripples. The crater is partly filled with aeolian basaltic sand. Detailed IDD work was done on Chocolate Hills. Pancam bands L2 (753 nm), L5 (535 nm), and L7 (432 nm) are shown as red, green, and blue colors.

Figure 15.

Pancam color view of the Chocolate Hills boulder showing fine-scale layering and a coating of dark material. MI views with Pancam color overlays (labeled “Super Res”) show the presence of hematitic concretions. Pancam image acquired on sol 2147 using bands L2 (753 nm), L5 (535 nm), and L7 (432 nm) as red, green, and blue colors, respectively.

Figure 16.

(a) Pancam enhanced false color image (RGB as 753 nm, 535 nm, 432 nm) of the Fe-Ni meteorite Block Island (∼65 cm wide). Box outlines areas shown in Figure 16b. (b) MI mosaic of upper portion of Block Island (∼7.5 cm across; merged with Pancam false color image) showing purple-hued coatings.

Figure 16.


[31] Chocolate Hills is an ejecta fragment from Concepción crater, a ∼10 m wide, relatively fresh impact crater located on the plains to the south of Victoria (Figures 2, 14, and 15). This rock is a finely layered, sulfate-rich material with hematitic concretions. The rock is partially coated with a mix of basaltic sand and hematitic concretions cemented by fine-grained hematite, based on analysis of MI, APXS, and MB data. The coating is interpreted to be a fracture filling deposit similar in origin to the fins found in Victoria crater and previous locations [Knoll et al., 2008]. Fins are indurated, fracture filling materials that are now raised features due to differential aeolian erosion of the surrounding softer sulfate-rich rocks. The presence of hematite implies that aqueous processes have been operative at least episodically since formation of the sulfate-rich sandstones that underlie Meridiani Planum.

[32] Bounce Rock was encountered just outside of Eagle crater and is rich in pyroxene, with a composition similar to the basaltic shergottite class of Martian meteorites [Zipfel et al., 2011]. A possible source for this ejecta fragment is the ∼20 km diameter Bopolu crater, located ∼75 km to the southwest of Eagle crater (Figure 1). The boulder Marquette Island was encountered during a traverse on the plains located to the south of Victoria crater (Table 2) [Mittlefehldt et al., 2010]. The mineralogy and composition of Marquette Island are similar to those of the Adirondack class of basalts examined by Spirit in Gusev crater. It probably originated as an ejecta fragment from an impact event that penetrated the Noachian crust, either beneath the sulfate-rich sedimentary deposits, or from the surrounding Noachian outcrops.

[33] Arkansas is a relatively small, dark cobble that was found close to Erebus crater. It is part of a group of cobbles found during traverses that were too small to brush or grind. Undisturbed surfaces, where large enough to be imaged with the MI, sometimes showed breccia-like textures. The composition and mineralogy of these cobbles indicate a mix of materials, including sulfates. Most likely, the Arkansas group of cobbles represents impact breccias strewn across the surface during local impact events and concentrated on the plains by rapid aeolian erosion of bedrock relative to the more highly indurated cobbles [Fleischer et al., 2010a].

[34] The cobbles of the Barberton group [Barberton (on the southern rim of Endurance), Santa Catarina (part of a strewn field on the rim of Victoria), Santorini, and Kasos] are chemically and mineralogically similar and thus probably have a similar origin [Schröder et al., 2010]. They are similar to mesosiderite meteorite silicate clasts, but may represent a group of meteorites not sampled on Earth. The largest accumulation of Barberton group cobbles, including Santa Catarina, is located near the rim of Victoria crater. It is possible that they are paired fragments of the impactor that formed that crater [Schröder et al., 2008; Ashley et al., 2009; Schröder et al., 2010].

[35] Opportunity has discovered three large (>35 cm) Fe-Ni meteorites (Block Island, Shelter Island, and Mackinac Island) south of Victoria crater [Fleischer et al., 2010b; Ashley et al., 2010]. Each exhibits discontinuous surface coatings that appear purple in Pancam false color images (e.g., Figure 16). For Heat Shield Rock, another Fe-Ni meteorite located to the south of Endurance crater, most RAT-brushed surfaces exhibit Pancam-derived spectra similar to laboratory spectra of the Canyon Diablo IAB meteorite (Figure 17). On the other hand, areas covered with purple coatings exhibit enhanced 535 nm band depths, and more negative spectral slopes between 753 nm and 934 nm, as compared to more typical natural or brushed surfaces on these meteorites (Figure 17). A nanophase iron oxide phase has, in fact, been identified in Mössbauer spectra associated with the purple coatings [Fleischer et al., 2010b]. Additionally, a minor, magnetically ordered iron oxide phase has been identified in spectra from Heat Shield Rock, inferring a small amount of relatively well-crystallized and larger particles (i.e., not nanoparticles) [Fleischer et al., 2010b]. In addition, APXS data for the Fe-Ni meteorites show elevated Br, Zn, and Mg values consistent with surface alteration.

Figure 17.

Pancam spectra (R*, relative reflectance normalized to cosine (incidence angle)) of Fe-Ni meteorites for (left) “typical” surfaces and (right) purple surfaces, compared to laboratory spectrum of Canyon Diablo convolved to Pancam band passes (offset by −0.1 for clarity). Spectra for Heat Shield Rock from Schröder et al. [2008] show surfaces pristine and brushed using the RAT. Pancam sequence identification numbers shown in legends. Error bars represent standard deviations of pixels selected for regions of interest in Pancam images. Canyon Diablo laboratory spectrum is RELAB MI-CMP-008, spectrum 001.

[36] Schröder et al. [2008] and Ashley et al. [2010] suggested that the coatings on the Fe-Ni meteorites represent remnants of a partially wind-eroded coating that formed when portions of the rock were buried, rather than a remnant fusion crust formed during traverse through the Martian atmosphere. These Fe-Ni meteorites have also undergone significant physical weathering (e.g., Mackinac Island has a cavernous interior) [Ashley et al., 2010]. The length of time that these meteorites have been on or near the surface is difficult to estimate. The size and mass of Heat Shield Rock and Block Island have been used as evidence that they landed during a period when the atmosphere was denser and slowed their descent [Beech and Coulson, 2010]. Otherwise they would have been destroyed during the subsequent hypervelocity impact with the surface. Alternatively, the landings would also have been possible with shallow entry angles under current atmospheric conditions [Chappelow and Golombek, 2010]. In any case, the extent of weathering of these Fe-Ni meteorites, combined with the presence of iron oxides in the stony-iron meteorites examined by Opportunity, will continue to provide insight into weathering processes on Mars.

[37] Observations of fresh craters younger than the north trending ripples (e.g., the Resolution crater cluster and Concepción) show abundant dark pebbles scattered across their surfaces. Given the apparent young ages of these craters, the straightforward explanation is that these dark pebbles are fragments of the impactors, suggesting that the widespread dark pebbles and cobbles observed by Opportunity at Meridiani Planum are lags of impactor-derived material (either meteoritic or secondary impactors from elsewhere on Mars) [Golombek et al., 2010].

7. Bedrock and Environments of Deposition and Alteration

[38] The sulfate-rich sandstones that comprise the Burns formation and examined by Opportunity within Eagle and Endurance craters provide compelling evidence of deposition by wind, with local subaqueous reworking within interdune ephemeral lakes [e.g., Squyres et al., 2004, 2006; Grotzinger et al., 2005]. Sulfate cements and hematitic concretions attest to multiple, but possibly short-lived episodes of percolation by acidic groundwaters [McLennan et al., 2005]. Recent calculations indicate that in situ iron oxidation could have provided sufficient acidity to explain the Burns formation mineralogy [Hurowitz et al., 2010]. Inferred grain compositions indicate that the sands were sourced in places where waters interacted with and weathered basaltic precursor rocks [Squyres et al., 2004; Squyres and Knoll, 2005]. During the period covered by this paper Opportunity explored outcrops on the plains and ventured into Erebus and Victoria craters to continue stratigraphic measurements designed to understand in more detail the origin and environments of deposition that produced the layered sulfate rocks that underlie the Meridiani plains (Figure 1). Particular emphasis was placed on the search for evidence of a sulfate-rich mud facies that might have been the source of the sandstones encountered by Opportunity. Finding those deposits would allow confirmation or rejection of the hypothesis that the sands were sourced in an evaporitic lake environment.

[39] The first set of very detailed measurements focused on the Olympia outcrops exposed to the northwest side of Erebus crater, together with a vertical section, dubbed Payson, on the southwestern wall of Erebus (Figures 1819). The ripple patterns in these outcrops provide compelling evidence for water transport of sulfate-rich sands, subsequently cemented to become sandstones [Grotzinger et al., 2006; Metz et al., 2009]. The Payson outcrop also showed disruption by water of laminated sandstones and the presence of shrinkage cracks, all consistent with an ephemeral shallow water environment. Compositional and mineralogical measurements acquired at the Olympia outcrops are very similar to measurements acquired in Eagle and Endurance craters. No in situ measurements were acquired at the Payson outcrops. The Olympia area is also the one place where it has proven possible to obtain in situ analyses of fin-like fracture fill, confirming that these features originated as clastic infillings of partings, later cemented to provide differential resistance to erosion. The fill is chemically similar to bedrock materials and not to modern soils. Taken together, the evidence suggests that these features formed after the primary phases of deposition and diagenesis, but prior to deposition of the modern soils [Knoll et al., 2008].

Figure 18.

HiRISE view of the highly degraded Erebus crater, with Opportunity traverses shown, along with representative sols and key targets. Opportunity conducted extensive remote sensing and IDD measurements at the Olympia outcrops (bright regions). For the Payson outcrop, which constitutes the southwestern wall of Erebus crater, systematic remote sensing was conducted while Opportunity traversed south toward Victoria crater. Note the extensive north-south trending ripples covering the crater and surrounding plains. Portion of HiRISE frame ESP_016644_1780_red.jp2.

Figure 19.

(a) Portion of a Navcam mosaic looking toward the west of a portion of the Payson outcrop acquired on sol 747. The outcrop height is ∼1.6 m and relatively dark ripples are shown in the plains beyond Payson. Note the outcrop cross bedding dipping toward the south. Box shows location of Pancam frame shown in Figure 19b. (b) Pancam view of a portion of the Payson outcrop showing approximately a dozen fine-scale, cross-bedded layers. The ripple patterns are indicative of shallow subaqueous transport, similar in interpretation to the ripple patterns in the sandstones observed to the north in the Olympia outcrop on the northwestern side of Erebus. Height covered in the image is ∼1.4 m. Pancam frame 1P194853277RSD646BP2547L7MZ acquired on sol 751.

Figure 19.


[40] Victoria is the largest crater examined by Opportunity to date, ∼750 m wide and ∼75 m deep. It was a primary target for exploration during the sols covered by this paper because of the extensive Burns formation stratigraphic exposures on its walls (Figures 2 and 20) [Squyres et al., 2009]. The approach to Victoria from the northwest allowed traversing across the annulus surrounding Victoria, a planar region that was found to consist of aeolian basaltic sands and hematitic concretions that partially cover the tops of beveled ejecta blocks (Figure 21). The ejecta deposit consists of relatively soft sulfate-rich rocks evenly eroded by wind to form the planar annulus that surrounds the crater [Grant et al., 2008]. Remote sensing of the crater wall from various promontories on the rim of Victoria showed that blocky ejecta deposits dominate the upper few meters of wall rock (Figure 22). The ejecta blocks are layered, contain hematitic concretions, and have coloration consistent with an origin as sulfate-rich bedrock. There is no evidence from Victoria's wall rocks or ejecta that the impact event penetrated into the underlying Noachian crust.

Figure 20.

HiRISE view of Victoria crater showing Opportunity's traverses, including drives into and out of Duck Bay for detailed IDD work on outcrops. The outcrop examined by Opportunity within Bottomless Bay is shown in a Pancam color view in Figure 22a. Traverses around a portion of Victoria's rim were conducted to map outcrops and to find a bay into which ingress and exit could be made with relatively low risk of embedding. Portion of HiRISE frame ESP_016644_1780_red.jp2.

Figure 21.

Pancam false color view of the Victoria ejecta deposit apron showing the tops of boulders that have been leveled by aeolian erosion and partially covered by soil with a relatively high concentration of hematitc concretions. The tear drop–shaped boulder on the top half of the frame is ∼0.9 m long. The target area is Malua, and the data were acquired on sol 1029. Pancam bands L2 (753 nm), L5 (535 nm), and L6 (482 nm) are shown as red, green, and blue colors.

Figure 22.

(a) Pancam false color view of the outcrop on the southwestern side of Bottomless Bay and a portion of the overlying ejecta deposit and annulus. Box shows region for which a geologic sketch map is shown in Figure 22b. Data acquired on sol 1037. Pancam bands 2 (0.753 μm), 5 (0.535 μm), and 7 (0.432 μm) are shown as red, green, and blue colors. (b) Geologic sketch map showing in-place bedrock, fractured bedrock, and poorly sorted ejecta blocks superimposed on the fracture bedrock surface. The ejecta has been leveled by aeolian erosion to form the annulus surrounding Victoria. Note rover tracks.

Figure 22.


[41] Duck Bay was chosen for entry into Victoria for detailed measurements because of the extensive Burns formation outcrops and the relatively easy ingress and exit paths (Figure 23). Beneath the ejecta deposit exposed at Duck Bay are four discrete layers that were examined using both remote sensing and in situ instrumentation [Squyres et al., 2009]. Steno is the layer in contact with the ejecta and is underlain by a relatively bright layer, Smith. Lyell and Gilbert are the next two layers examined during the Duck Bay campaign. Steno consists of a fine to medium-grained sandstone with well-defined laminae. Cross bedding is evident, as are hematitic concretions. Steno is separated from Smith by an unconformity. Smith is a relatively bright sandstone and exhibits fine-scale laminations. Lyell is transitional with Smith and exhibits tabular, prismatic vugs, cross bedding, and an abundance of hematitic concretions. Gilbert was only measured in one location and the contact with Lyell is gradational. Lyell and Smith are also sandstones. Pancam observations show that the Smith unit has an abrupt spectral downturn at 1000 nm, consistent with the presence of the molecular water vibrational mode 2v1 + v3 and 3vOH for OH-bearing minerals (Figure 23) [see Rice et al., 2011]. No evidence was found in any of the layers for the mud facies that might have been the source for the sulfate-rich sandstones. In fact, the rocks examined in Victoria, both using remote sensing from capes, and detailed measurements in Duck Bay, are best interpreted as sulfate-rich aeolian sands altered and cemented by groundwater infiltration [Squyres et al., 2009].

Figure 23.

Portion of a Pancam false color mosaic covering the stratigraphic section examined by Opportunity during its traverses within Duck Bay, Victoria crater. The mosaic data were acquired between sols 970 to 991 as part of the Cape Verde panorama. Steno is the topmost in-place outcrop beneath the ejecta deposit. The brighter layer, Smith, can be found in many locations around the perimeter of the crater. Pancam 13F spectra for Smith, Lyell, and Gilbert are shown in the lower right and discussed in detail in the text. A hydration index based in the depth of the 1.0 μm band is shown on the lower left and indicates that the Smith unit is hydrated. The hydration index covers the left portion of the Pancam false color mosaic and is centered vertically on the Smith unit. Pancam bands 2 (753 nm), 5 (535 nm), and 7 (432 nm) are shown as red, green, and blue colors.

[42] In situ data were acquired for undisturbed, brushed, and ratted rock targets within each of the four stratigraphic layers in Duck Bay. Correspondence analysis shows the importance of removing aeolian sand and dust covers and any coatings from these rocks to understand their intrinsic characteristics (Figure 24). In particular, the first factor in APXS data, accounting for 92% of the variance of the data set, shows a trend from basaltic to more sulfate-rich materials for natural, brushed, as opposed to ratted targets. Ratted targets have the highest sulfur content and least contamination by coatings or basaltic sands. The second factor, accounting for 4% of the variance, shows that the ratted rock targets can be discriminated from one another on the basis of chlorine content, with Gilbert showing the highest value, and Steno the lowest. A trend of increasing chlorine content with increasing depth is also evident in a scatterplot of chlorine to silica content as a function of depth (Figure 25). On the other hand, the sulfur and magnesium contents, relative to silica, both decrease as a function of depth beneath the surface (Figure 26). These compositional patterns correlate well with the hydration index computed from the depth of the 1000 nm band evident in the Smith unit (Figure 24). Further, this bright, upper unit appears to continue around the entire crater and can be fit with a horizontal plane (A. Hayes et al., Reconstruction of eolian bedforms and paleocurrents from cross-bedded strata at Victoria crater, Meridiani Planum, Mars, submitted to Journal of Geophysical Research, 2011).

Figure 24.

Correspondence analysis plot for APXS data acquired for rock outcrops in Duck Bay, Victoria crater. The highest fractional variance is controlled by changes in chemistry from natural surfaces, contaminated by aeolian basaltic soils and coatings, to ratted surfaces that better represent the sulfate-rich outcrop chemistry. Brushed targets are denoted by “_b” after names. Targets in italics have been ratted (“_r”), and numbers in parentheses correspond to ferric iron to total iron values from MB observations. The direction of second highest fractional variance separates ratted targets based on chlorine, sulfur, and magnesium contents, as shown by scatterplots in Figures 25 and 26. Box in the top right corner is a schematic stratigraphic section.

Figure 25.

APXS-based chlorine/SiO2 values are shown as a function of depth beneath the ejecta to bedrock contact for ratted targets in Victoria and Endurance craters. The presence of the bright upper layer in both craters and the increase in Cl/SiO2 with increasing depth indicates a regional-scale aqueous process that concentrated relatively soluble Cl in lower stratigraphic horizons. The bright layer corresponds to the Smith unit in Victoria and to rock targets above the Whatanga contact for the Endurance Karatepe section. Endurance targets Virginia, Kentucky, Ontario, and Tennessee are above the Whatanga contact.

Figure 26.

Decreasing values of (top) MgO and (bottom) SO3 relative to SiO2 with depth beneath the ejecta to bedrock contact for both Endurance and Victoria craters, implying a regional-scale aqueous alteration process that led to an enrichment of relatively insoluble magnesium sulfates in the bedrock upper layers.

Figure 26.


[43] The rocks examined in Duck Bay, although still part of the Burns formation, are separated laterally and topographically from the Karatepe section outcrops examined in Endurance crater. Even so, the Karatepe section also shows a bright upper layer (above the Whatanga contact) that is depleted in chlorine relative to silica and enhanced in magnesium and sulfur relative to silica as compared to rocks exposed at greater depths (Figures 2526) [see also Squyres et al., 2009]. In addition, the VNIR multispectral character of both sets of strata is similar [Farrand et al., 2007]. A. Hayes et al. (submitted manuscript, 2011), examining orbital data, found that bright layers are evident in a number of other craters that formed in the Burns formation. Overall, the evidence is interpreted to reflect regional-scale differential vertical mobility of soluble sulfate and chloride salts during near surface aqueous-mediated diagenesis [Clark et al., 2005; Amundson et al., 2008]. The observation that Victoria ejecta deposits include fragments of the Smith unit implies that the aqueous alteration event predates formation of Victoria [Edgar et al., 2010].

8. Rim of Endeavour Crater and Adjacent Layered Sedimentary Rocks

[44] Opportunity has thus far been exploring sedimentary rocks and soils that unconformably overlie the Noachian crust (Figure 1). Endeavour crater predates the sedimentary deposits and the crater rim exposes materials of Noachian age (Figures 1, 27, and 28). This is clear from geologic mapping and also initial spectral analysis of CRISM hyperspectral data [Murchie et al., 2007] covering the rim and surrounding areas. Specifically, Wray et al. [2009] showed from analysis of CRISM spectra covering 0.4 to 2.5 μm in wavelength that portions of the rim expose iron and magnesium-rich smectite clay minerals. In addition, these authors showed that the sedimentary rocks adjacent to the rim have spectra that are indicative of hydrated sulfates.

Figure 27.

CTX mosaic with Opportunity traverses shown and locations on the rim of Endeavour labeled. The white lines extending from the sol 2239 position cover the field of view of the Pancam superresolution images of the rim of Endeavour shown in Figure 28. Mosaic generated from CTX frames P13_006135_1789_XN_01S005W_071117, P15_006847_1770_XN_03S005W_080111, and P17_007849_1793_XN_00S005W_080330.

Figure 28.

Pancam superresolution view of a portion of Endeavour's rim and Iazu ejecta acquired from ∼13 km distance. Cape Tribulation has exposures of Fe-Mg smectite clay minerals based on analyses of CRISM data [Wray et al., 2009]. Analysis of CRISM data also indicate the presence of layered sedimentary rocks with hydrated sulfate signatures adjacent to the rim [Wray et al., 2009]. This image was generated from a series of eight Pancam frames acquired on sol 2239.

[45] The long-term objectives for the Opportunity extended mission are to drive to the hydrated sulfate deposits and Noachian-aged rim materials of Endeavour. By sol 2239 the rover was within ∼11 km of the rim (Figure 27). Pancam color and superresolution imaging (Figure 28), combined with periodic collection of imaging data from the HiRISE, CTX, and CRISM instruments, are helping to define traverses to locations where the hydrated sulfate sedimentary rocks and altered rim materials are best exposed and accessible to Opportunity. Outcrops of the Burns formation that have been characterized thus far by Opportunity exhibit OMEGA-based and CRISM-based reflectance spectra that indicate the spectral dominance of relatively anhydrous phases [e.g., Arvidson et al., 2006]. This result is consistent with the dominance of nanophase iron oxide coatings on rock surfaces [e.g., Knoll et al., 2008], and with the observation that the 6 μm bending vibration for water is not evident in Mini-TES for undisturbed rock surfaces [Glotch et al., 2006]. On the other hand, Mini-TES deconvolution of ratted Burns formation material, including constraints from MB and APXS measurements, indicate the presence of hydrated Mg and Ca-sulfate minerals [Glotch et al., 2006]. The surface exposures of hydrated sulfates close to the rim of Endeavour are likely layers that lie stratigraphically beneath the Burns formation rocks examined by Opportunity. Characterizing the composition, mineralogy, and texture of these older sedimentary rocks will provide new information on paleoenvironmental conditions and perhaps even provide the evidence for the source rocks for the sulfate-rich aeolian sandstones that dominate the Burns formation. In addition, Opportunity's characterization of Endeavour's rim rocks, including clay minerals, will allow even older environmental conditions to be reconstructed.

9. Conclusions

[46] Opportunity has been traversing across the plains of Meridiani since January 2004, far exceeding the expected lifetimes and traverse distances of the rover, and using its Athena scientific payload for many more measurements than originally planned. Opportunity has operated over three Martian years and acquired important information on modern atmospheric dynamics, including atmospheric opacity, clouds, and use of atmospheric argon as a tracer for circulation dynamics. The aeolian ripples traversed by Opportunity were generated by easterly winds in an ancient environment, probably within hundreds of thousands of years, when the spin axis obliquity was higher and Hadley cell circulation enhanced. Cobbles and boulders examined by Opportunity include local and regional-scale ejecta blocks, together with both stony iron and iron-nickel meteorites. The meteorites have undergone both physical and chemical weathering and are likely a lag of impactor-derived materials. Opportunity has made measurements within Eagle, Endurance, Erebus, and Victoria craters, together with outcrop exposures on the plains that were focused on characterizing the formation and modification of the Burns formation sulfate-rich sandstones. Results continue to show compelling evidence of sand deposition by wind, with local reworking within ephemeral lakes. Extensive lacustrine evaporitic facies have not yet been found, although particular emphasis has been placed on finding these putative materials. Accessing the hydrated sulfate rocks near the Endeavour crater rim and the clay minerals on the rim proper will open a new chapter for Opportunity and allow characterization of materials not yet encountered during the mission.


[47] We thank the capable team of engineers and scientists at the Jet Propulsion Laboratory and elsewhere who made the Opportunity mission possible. We also thank support from NASA for the MER science team to allow both collection and analysis of data from Opportunity. Alejandro Soto and Mark Richardson provided valuable comments on an earlier draft of this paper and we thank them for their efforts. Thanks to Emma Reinemann for helping with several figures.