Mineralogy of recent volcanic plains in the Tharsis region, Mars, and implications for platy-ridged flow composition

Earth and Planetary Science Letters xxx (2009) xxx–xxx

EPSL-09947; No of Pages 11

Contents lists available at ScienceDirect

Earth and Planetary Science Letters

j ourna l homepage: www.e lsev ie r.com/ locate /eps l

ARTICLE IN PRESS

Mineralogy of recent volcanic plains in the Tharsis region, Mars, and implications forplaty-ridged flow composition

N. Mangold a,b,⁎, D. Loizeau b,c, F. Poulet c, V. Ansan a, D. Baratoux d, S. LeMouelic a, J.M. Bardintzeff b,B. Platevoet b, M. Toplis d, P. Pinet d, Ph. Masson b, J.P. Bibring c, B. Gondet c, Y. Langevin c, G. Neukum e

a Lab. Planétologie et Géodynamique de Nantes, CNRS/Univ. Nantes, Franceb Interactions et Dynamique des Environnements de Surface, UMR8148, CNRS/Univ. Paris Sud, Francec Institut d'Astrophysique Spatiale, CNRS/Univ. Paris Sud, Orsay, Franced Lab. Dynamique Terrestre et Planétaire, OMP, CNRS/Univ. Toulouse, Francee Institut of Geosciences, Freie Universität, Berlin, Germany

⁎ Corresponding author. Lab. Planétologie et GéodynamNantes, 2 rue de la Houssinière, 44322 NANTES Cedex,

E-mail address: [email protected] (N.

0012-821X/$ – see front matter © 2009 Elsevier B.V. Adoi:10.1016/j.epsl.2009.07.036

Please cite this article as: Mangold, N., et aridged flow composition, Earth Planet. Sci.

a b s t r a c t
a r t i c l e i n f o
Article history:Accepted 28 July 2009Available online xxxx

Editor: T. Spohn

Keywords:Marsvolcanismplaty flowspyroxenes

Volcanism on Mars occurred until recently, but the mineralogy of recent lava plains is poorly known becausefew regions display fresh outcrops devoid of dust. Using visible and near infrared data of the Mars Expressprobe, two new volcanic plains in Noctis Labyrinthus have been identified, and the existence of a volcanicplain on the floor of Echus Chasma has been confirmed. Crater retention ages estimated for these three plainsrange between 50 and 100 My, corresponding to the Late Amazonian. These plains represent an excellentopportunity to constrain the mineralogy of recent volcanic rocks. Results show that basaltic compositionswith plagioclase and high calcium pyroxene are predominant. The low olivine proportion suggests that theapparent fluidity of these flat plains is not related to magmas being ultramafic. In addition, a platy-ridgedtexture is observed in two of the studied regions. Our study shows, for the first time, that this texture isassociated with volcanic rocks, and that these rocks are of typical basaltic mineralogy. Finally, these volcanicplains are located more than 1000 km east of previously known Late Amazonian volcanic centers of theTharsis region, an observation to be taken into account when considering models of recent volcanism onMars.

© 2009 Elsevier B.V. All rights reserved.

1. Introduction

Volcanism onMars is diverse andwell developed, with evidence ofabundant plain-style volcanism and a variety of edifices. At the scaleof Viking images, early investigators observed that the most recentvolcanism was located in distinct areas such as the Tharsis (includingOlympus Mons) and Elysium regions (e.g., Greeley and Spudis, 1981).At that time, it was unclear if volcanic activity extended into the LateAmazonian (b500 My). A better picture of recent volcanism is nowgiven by the Mars Observer Camera (MOC) onboard Mars GlobalSurveyor (Hartmann and Neukum, 2001; Berman and Hartmann,2002) and by the High Resolution Stereo Camera (HRSC) onboardMars Express (Neukum et al., 2004; Chapman et al., 2007; Vaucheret al., 2009; Werner, 2009) (e.g., Arsia and Olympus Montes). Ages asrecent as several tens of millions of years have also been deduced forthe Central Elysium Planitia, southeast of ElysiumMons (e.g., Vaucheret al., 2009). These young ages are of interest because they show thatvolcanism on Mars may still be active. This study focuses on three

ique de Nantes, CNRS et Univ.France.Mangold).

ll rights reserved.

l., Mineralogy of recent volcLett. (2009), doi:10.1016/j.e

plains, of which two are identified here for the first time using HRSCimagery.

High resolution imagery and topography can be combined withvisible and near infrared spectral data to study both morphology andcomposition of volcanic rocks onMars (e.g., on SyrtisMajor, Poulet et al.,2003). However, few spectral data exist for areas of recent volcanism,because suchsurfaces are often featurelessdue towidespreaddust cover(e.g., Hamilton et al., 2003, Stockstill-Cahill et al., 2008). As described inthis manuscript, three small regions (b50 kmwide)were found to haveformed recently (b100 My). Two of these areas were found in NoctisLabyrinthus, and one was found on the floor of Echus Chasma (Fig. 1).The resolution of OMEGA (Observatoire pour la Minéralogie, l'Eau, lesGlaces et l'Activité) spectral data provides the possibility to analyze thecomposition of these recent plains in detail (Bibring et al., 2004).

Plain-style volcanism, corresponding to broad lava flows withoutobvious volcanic vents, is widespread on Mars (e.g., Hodges andMoore, 1994). However, few mineralogical data exist with which toconstrain their composition independently from rheological evidencesuggesting a high fluidity. Additionally, platy-ridged flows that arecommon in Martian plains are the subject of debate due to theirmorphologic similarity with potential ice rafts, suggesting frozen seasin the southern Elysium plains (Murray et al., 2005). Two of the three

anic plains in the Tharsis region, Mars, and implications for platy-psl.2009.07.036

mailto:[email protected]

http://dx.doi.org/10.1016/j.epsl.2009.07.036

http://www.sciencedirect.com/science/journal/0012821X


Fig. 1. (top) Context map: MOLA shaded relief of the Tharsis plateau. (bottom) TESthermal inertia of the area in a extracted from Putzig et al. (2005). The three studiedareas, NL1, NL2, and ECF, display inertia N300 J m−2 K−1 s−1/2 (in yellow), while mostof the volcanoes and lava plains are buried beneath low inertia material (b200 J m−2

K−1 s−1/2). (For interpretation of the references to colour in this figure legend, thereader is referred to the web version of this article.)

2 N. Mangold et al. / Earth and Planetary Science Letters xxx (2009) xxx–xxx

ARTICLE IN PRESS

plains studied showed platy-ridged landforms with fresh outcropsenabling us to extract their mineralogy. Thus, our study focuses on themorphology, mineralogy of recent volcanic plains that represent anexcellent opportunity to constrain the composition of recent volcanicrocks and understand their apparent fluidity, in order to provideinsights for the origin of recent volcanism on Mars.

Table 1OMEGA derived modal mineralogy.

HCP LCP LCP/HCP Neutral Olivine Dust RMS

Echus Chasma 25±6 6±3 23±13 47±10 b5 17±13 0.26Noctis Labyrinthus 1 34±5 5±3 17±10 60±10 b5 b1 0.39Noctis Labyrinthus 2 30±5 8±3 29±15 48±7 b5 12±9 0.39

Values are percentage (including the LCP/HCP ratio) and standard errors forabundances indicate ±1σ. Neutral end-member corresponds to plagioclases.

2. Methods

2.1. HRSC/Mars Express data

The HRSC camera acquired images in five panchromatic channelsunder different observation angles, as well as four color channels at arelatively high spatial resolution (Neukum and Jaumann, 2004). Inour work we used only panchromatic nadir images, with a maximumspatial resolution from 10 to 40 m pixel−1, and two panchromaticstereoscopic images with a spatial resolution usually degradedcompared to the nadir image. The coordinates of ortho-rectifiednadir images are defined in the planetocentric system of the MarsIAU 2000 ellipsoid (Seidelmann et al., 2002). The following imageshave been processed: #1977, #1999, #2402, #2479, and #3155 inthe Noctis region, with a spatial sampling of the nadir image at 15 mpixel−1 for #1977, #1999, and #3155, and 40 m pixel−1 for the#2402 and #2479. Images of orbit #71 and #5379 are used in theEchus Chasma region with a spatial sampling at 30 m pixel−1.

Please cite this article as: Mangold, N., et al., Mineralogy of recent volcridged flow composition, Earth Planet. Sci. Lett. (2009), doi:10.1016/j.e

HRSC DEMs (Digital Elevation Models) were computed using thephotogrammetric software developed at the DLR and the TechnicalUniversity of Berlin (Scholten et al., 2005; Gwinner et al., 2005, Ansan etal., 2008). The image correlationwas performedusing amatching processat a different spatial grid size (Scholten et al., 2005). The height wascalculated taking into account the Martian geoid, defined as thetopographic reference for Martian heights, i.e. the areoid (Smith et al.,1999). In theNoctis Labyrinthus region, theDEMswere constructedusingspatial griddingof ~30m/pixel fororbits#1977and#1999, and100mfororbit #2479. In the Echus Chasma region, only theMOLA (Mars ObserverLaser Altimeter) DEM at 1/128 degree was used for the topography.

2.2. OMEGA/Mars Express data

OMEGA is a visible and near infrared (VNIR) hyperspectral imagerproviding three-dimensional data cubes with spatial samplings from afewkilometers to 300m. For each pixel it provides spectra between 0.35and 5.1 µm, using 352 contiguous spectral elements (spectels), 7–20nmwide. The spectrometer consists of three detectors (from 0.35 to 1 µm,from0.9 to 2.7 µm, and from2.5 to 5.1 µm) (Bibring et al., 2004). For thisstudy we used data recorded by the second detector where the sig-natures of the mafic minerals are well characterized. Pyroxene isdetected from the presence of a broad band at 1.9 and 2.3 µm, res-pectively, for low-calcium pyroxene (LCP) and high calcium pyroxene(HCP). In our study, the detection of pyroxene is done with the spectralindex used by Poulet et al. (2007) sensitive to the presence of both HCPand LCP. To account for instrumental and atmospheric biases, thedetection is considered positive for values of the pyroxene spectralparameter larger than 1% (Poulet et al., 2007). The detection of olivine isbased on the broad 1 µm band from the spectral parameter defined inMustard et al. (2005), but no olivine has been detected by this methodon the region studied. The depths of absorption bandswere determinedand mapped, allowing the presence of minerals to be identified in thetop few micrometers of the Martian surface.

The identification ofminerals from band depths does not, however,provide a direct determination of all minerals. In addition, the spectralparameter used for pyroxene is not designed to discriminate betweenthe LCP end-member and the HCP end-member. Modeling of NIRspectra has been demonstrated to provide an accurate estimate formineral abundances in mixtures of basaltic granular materials for awide range of particle sizes (Poulet and Erard, 2004). As described byPoulet et al. (2009a), scattering models may be used to explore theparameter space and the influence of grain size of mineralogicalmembers, including spectrally neutral components. The model hastwo free parameters for each end-member: the average grain size andthe relative abundance. Since aerosols influence spectral slopes, oneadditional free parameter is used to adjust the spectral slope con-tinuum. The spectra are fitted in the 0.99–2.49 µm wavelength range(OMEGA SWIR-C channel) using a simplex minimization algorithm.The synthetic spectrum is a nonlinear combination of the opticalindices of theminerals selected to be part of themixture, in proportionto their abundance. Upon obtaining a data spectrum fit, the algorithmsupplies the user with a model-derived spectrum as well as with apercentage and grain size for each end-member used in the fit. Theprincipal mineralogical characteristics of the three areas studied weredetermined using this method and are summarized in Table 1.



Fig. 2. OMEGA spectral mapping of pyroxenes in Noctis Labyrinthus. A map of HCP superimposed on a THEMIS daytime mosaic of the Noctis Labyrinthus region. Squares representthe two regions of interest NL1 (west) and NL2 (east). Colors represent band depths from 1% (dark blue) to 5% (red) for both maps. Mars Express orbits used are: #305, #331, #357,#431, #453, #486, #519, #1052, #1107, #1933, #1944, #1955, #2336, #2347, #2369, #2380, #2402, #2479, #3078, #4219, #4230, #4241, #4545, and #4556. (For interpretation ofthe references to colour in this figure legend, the reader is referred to the web version of this article.)

3N. Mangold et al. / Earth and Planetary Science Letters xxx (2009) xxx–xxx

ARTICLE IN PRESS

2.3. Other datasets

MarsObserver Camera (MOC) images are used to complementHRSCnadir images at a better resolution, typically 2.8 m pixel−1 (Malin et al.,1992). Nighttime and daytime infrared images of the Thermal EmissionImaging System (THEMIS) onboardMars Odyssey were used to providea view of the surface properties (Christensen et al., 2003). Only themosaics of thermal images are used at 230mpixel−1.We chose to use acolor scale for these images and superposed themonto HRSC or THEMISdaytime images in order to observe where thermal images may beindicativeof thepresence of indurated rockoutcropsor coarsegrains (asseen in red for high nighttime temperatures) compared to more dusty,less indurated particles (as seen in blue for low night time tempera-tures). Values of thermal inertia for the regions of interest have beenextracted from the TES (Thermal Emission Spectrometer) thermalinertia map at a resolution of 3 km (Putzig et al., 2005).

3. The floor of Noctis Labyrinthus Chasmata

3.1. Regional context and the distribution of pyroxenes as seenby OMEGA

Pyroxene signatures are present throughout all of the southernpart of Tharsis and in Valles Marineris (Fig. 2). Hence, these regionsare devoid of dust as seen on the thermal inertia map (Fig. 1).However, most pyroxene signatures are observed either on dark sanddunes filling canyons floor, such as in Candor Chasma (Mangold et al.,2008), or on canyon walls (e.g. Mustard et al., 2005). The source ofmaterial is unknown for canyon floors. For canyon walls, rockscorrespond to thick accumulations of lava flows of Hesperian orNoachian age, which are beyond the scope of our study. By examiningthe region of Noctis Labyrinthus (Fig. 2), we observe pronouncedpyroxene signatures in the west of Syria Planum and West TithoniumChasma.


Syria Planum is a volcanic plain with many small low shieldvolcanoes of Hesperian age (Baptista et al., 2008). To the east, WestTithonium Chasma is covered extensively by sand sheets and dunesthat bury the canyon, and part of the Oudemans crater (the 100 kmdiameter crater to the southeast of the canyon). Therefore, none ofthese signatures have been found to be correlated with youngvolcanic rocks. However, our interest focused on two spots (NL1and NL2 on Fig. 2) inside Noctis Labyrinthus (Fig. 2), where pyroxenesignatures appear to be pronounced and the thermal inertia is high.Spectra of both regions (Fig. 3) fit well with typical laboratory spectraof pyroxenes, especially HCP, with broad 1 µm and 2.3 µm bands.

3.2. West canyon floor (NL1)

The first canyon floor analyzed, NL1, is located at 99W, 7S (Fig. 4).It is about 50 km from west to east, 60 km from north to south, and5 km deep below plateau level. The HRSC image shows a smooth flooronly interrupted by small buttes. The albedo of the floor is very lowcompared to walls and plateaus (Fig. 4a), and low albedo correlateswith the high night time temperatures seen on the floor (Fig. 4d),corresponding to areas with more than 500 J m−2 K−1 s−1/2 on theTES thermal inertia map (Putzig et al., 2005) (Fig. 1b). Only theuppermost part of the canyon walls (rock scarps) display nighttimetemperatures as high as the canyon floor (Fig. 4d). Pyroxenesignatures detected by OMEGA are found only on the canyon floor.The topography of the floor measured by MOLA data (Fig. 4b,c)indicates that the floor is at a constant altitude of ~2250 m withlimited variation (b50 m). All of the pyroxene signatures (Fig. 4e) aredetected beneath this elevation contour.

A closer look to the east of this canyon is possible with moreresolved HRSC data from orbit 1999 (Fig. 5). Here, the HRSC DEMallows us to observe that all small buttes located on the canyon floorsare located above the 2250 m contour. This contour separates thatpart of the floor with low albedo (0.15 at 1 µm OMEGA reflectance)



Fig. 3. Spectra (average of 9 pixels) of the selected areas NL1, NL2 and ECF. Libraryreflectance spectra are relative reflectance. Library spectra correspond to diopside (HCP)and pigeonite (LCP) from the RELAB database (Reflectance Experiment LABoratory;http://www.planetary.brown.edu/relabdata/). A spectrum from the dust region aroundNL1 is shown for comparison.

Fig. 4. Close-up on NL1: (a) HRSC image #2379 inside the NL1 trough. (b) Same image as a w2300m showing the very flat Chasma floor. (d) The HRSC imagewith THEMIS nighttimemoslow temperatures, and, therefore, finer material than in warmer locations. (e) An OMEGA mreferences to colour in this figure legend, the reader is referred to the web version of this a


ARTICLE IN PRESS


relative to surrounding bright terrain of smooth texture, containingpyroxene. These observations confirm that the whole floor is very flatat the elevation of 2200 to 2250 m over the entire canyon floor area.

Close-ups of the canyon floor reveal further landforms not visibleat a broader scale. First, a circular shape 1 km in diameter and about100 m in depth is observed beyond the southern edge of the plain(Fig. 6a,b). It is breached in its northern part by a 200 m large, 30 mdeep channel which joins the canyon floor 5 km to the north andvanishes inside the canyon floor shortly after intersecting the 2250 melevation contour (Fig. 6b). In addition, most of the canyon floor issmooth at the HRSC scale with only a few sand dunes present in thecentral part. Nevertheless, the MOC close-up shows small impactcraters and local outcrops, especially around unfilled pits (Fig. 6c),suggesting that sand exists but does not mantle all the surface. Theseimages also display a lineation of NW–SE direction, resembling aburied dyke or a fracture zone, with collapse pits 100 m wide sittingon the fracture (Fig. 6c). Finally, the area located east of the canyonfloor exhibits not only small buttes, but a series of pitted cones ofabout 100 m in diameter, which are observed at the contact of the2250 m altitude contour (Fig. 6d).

All these observations listed above indicate a volcanic origin forthe canyon floor. By itself, the pyroxene-rich mineralogy coupled withthe low albedo and the strong thermal inertia is good evidence for avolcanic origin (Fig. 4). Additionally, the combination of the deepcircular landform together with a channel (Fig. 6) is similar to ter-restrial volcanic vents with lava channels emerging from them (e.g.pit crater with lava channel in the Snake River plain, Idaho, USA, Fig.5G, in Hodges and Moore, 1994). Pitted fractures inside the canyonfloor might correspond to other vents, which were partially filled bythe lavas they extracted. The pitted cones to the east might be anadditional source of lavas. Alternatively, they might correspond torootless cones, also named pseudocraters. Rootless cones form at thecontact of lava flows and a volatile such as liquid water or water ice(e.g. Lanagan et al., 2001). Their presence at the contact with the

ith MOLA contours (250m intervals). (c) MOLA DEM in between elevations 2200 m andaic superimposed in color. Red colors represent high temperatures, blue colors representap of pyroxene superimposed on the THEMIS daytime image. (For interpretation of therticle.)


http://www.planetary.brown.edu/relabdata/


Fig. 5. (a) An HRSC close-up of orbit #1999 on the eastern side of the Chasma floor. (b)The HRSC DEM represented in between 2200 m and 2300 m of elevation, which showsthe flat floor to the west and the presence of hills above the floor to the east.

Fig. 6. (a) HRSC orbit #1999 close-up of the southern part of the Chasma. (b) The sameas a with HRSC DEM contours superimposed. This area displays a central pit to the southfrom which emerges a sinuous channel ending in the Chasma floor. (c) MOC image#R1800704 in the central area showing two elliptical pits formed along a fracture zone.(d) An HRSC close-up on the eastern margin of the Chasma floor showing a series ofsmall (b500 m) cones in the center, and round hills that emerge from the floor of theChasma by a few tens of meters at the maximum.


ARTICLE IN PRESS

canyon floor is consistent with such a hypothesis. The apparent size ofthese cones do match terrestrial cones as well as some other Martianexamples (Lanagan et al., 2001; Meresse et al., 2008) andmay indicatethe presence of liquid water or water ice at the time of the volcanicactivity. Finally, the constant altitude of the canyon floor suggests thatthe volcanic material is fluid enough to form a smooth plain whichfollows an equipotential surface. All this evidence favors formation asa lava plain, covering about 450 km2 if the 2250 m contour is taken asa reference.

3.3. East canyon floor

The second canyon floor analyzed, NL2, is located at 96.2W, 7.2S(Fig. 7). It is about 50 km in length from west to east, 40 km in lengthfrom north to south, and 6 km in depth. The HRSC image shows asmooth floor with small buttes in the eastern and southern part. Thealbedo of the floor is low when compared with the walls and plateaus(0.11 at 1 µm OMEGA reflectance), and displays a gradation throughthe floor from the west to the east, the eastern region being darkerthan the western region (Fig. 7a). Low albedo correlates with the highnight time temperatures seen on the canyon floor (Fig. 7d), reachingup to 600 J m−2 K−1 s−1/2 on TES data in the darkest areas (Putziget al., 2005) (Fig. 1b). Here too, only the uppermost part of canyonwalls (namely the rock scarps) display temperatures as high as thecanyon floor. Pyroxene signatures detected by OMEGA are foundmainly on the eastern part of the canyon floor (Fig. 7e).

The topography of the floor as measured by HRSC data (Fig. 7b,c)indicates that the floor is constrained between the altitudes of 1850and 2000 m. The 2000 m contour follows the smooth texture of thecanyon floor very well. We observe that the transition between hightemperatures (red) and low temperatures (blue to yellow) at thesouthern and northern edge of the canyon floor correlates with the2000 m altitude level (Fig. 7d). Dust, as seen from blue colors (lownight temperatures) and high albedos in the HRSC image, mantlesmost of the canyon walls and debris aprons. As dust deposition is agradual process at the regional scale, this observation likely illustratesthat dust was deposited before the canyon floor was resurfaced, andthat dust was buried beneath the high thermal inertia material. Even


so, the fact that the albedo is slightly higher in the western part of thecanyon floor suggests that the western part has been slightly mantledby a thin (b10 cm) coating of fresh dust hiding the rock surface atOMEGA wavelengths. An elliptical feature, marked by “E” is visible inthe albedo (Fig. 7a), in the topography (Fig. 7c), and in thetemperature maps (Fig. 7d). This landform is oriented NW–SE, is afew tens of meters deeper than the rest of the canyon floor, and couldhave served as a trap for recent dust deposition.

A close-up on HRSC andMOC data shows a set of different textures(Fig. 8). First, the southern part exhibits a specific texture that differsfrom the smooth canyon floor. Here, plains appear as if it has beendisrupted by liquefaction, or some kind of fluid activity. Indeed, smallpieces of plains are found in themiddle of this area and resemble solidplates floating over a fluid medium (Fig. 8a). The plates are still assmooth as the plains material, as seen on the MOC close-up (Fig. 8b).In between these plates, the texture is rough with a lot of fracturesthat do not cross the plates. Plates are up to 1 km long and are lessthan 30m thick because the resolution of the HRSC DEM is insufficientto detect the plates.

Outside this area, plains exhibit frequent small ridges (marked by“R” in Fig. 8c) several kilometers long and approximately 100 mwide.The ridges display a positive relief, as if they were formed bycompression in soft material, but their curved shape without straightfaults (Fig. 8c) does not plead for a tectonic origin by brittle faulting. InFig. 8c, left of the ridge, curvilinear shapes of material that appear to



Fig. 7. Close-up on the NL2 trough: (a) HRSC image #1977 inside the NL1 trough. (b) The same as a with HRSC DEM contours (250 m intervals). Note the presence of irregularcontours for the 2000 m elevation contour, which is due to the DEM accuracy (b30 m) relative to the flatness of the area. (c) An HRSC DEM in between elevations 1860 and 2080 mshowing the very flat Chasma floor with a possible elliptical trough in themiddle (identified by E in a). (d) The HRSC image with the THEMIS nighttimemosaic superimposed in color.(e) An OMEGA map of pyroxene superimposed on the THEMIS daytime image.


ARTICLE IN PRESS

have been deformed in a ductile way can be observed. Small impactcraters can also be observed in this image.

In summary, this volcanic plain has a relatively high nighttimetemperature with rough textures of bedrock exposure with lowalbedos and pyroxene signatures on fresh outcrops. These character-istics argue in favor of a volcanic origin such as lava flood plains. Theplate shape, size and width are consistent with those frequentlypresent in Martian lava flood plains, usually referred to as platy-ridged lava flows (e.g., Keszthelyi et al., 2000). Other landforms suchas ridges and curved textures are also consistent with viscous lavaflows, despite the fact that this type of ridge does not seem frequent inother Martian flood plains. The total surface of this volcanic plain isabout 800 km2. A difference between the NL2 and NL1 plains is thatthere is no obvious vent found in, or along the plain NL2. Nevertheless,it is possible that the elliptical trough (E) corresponds to a lava ventthat has been entirely buried beneath its flows.

4. Echus Chasma floor

Echus Chasma is the source area of the Kasei Vallis outflowchannels that dissect volcanic plateaus from the equator to 20N, westof the large volcanoes of Tharsis (Fig. 1). The Echus-Kasei valley ispartially filled by lava flows originating from the west, from the mainvolcanic center (e.g. Scott and Tanaka, 1986). While no vent isobserved in our studied area, volcanic fissures exist on the floor of thevalley north of the area studied, suggesting that local emissions fromthe Chasma floor are also possible (Chapman et al., submitted forpublication).

The studied area is a zone with two types of material, a highthermal inertia, low albedo region, in the middle of plains that aregenerally of low thermal inertia and high albedo (Fig. 9). Pyroxene isdetected where the albedo is low (0.19 in 1 µm OMEGA reflectance)and the inertia high (Fig. 9d, e). The morphology observed in theseplains shows a lack of sand dunes and a rough surface at MOC scalesuggesting the surface is relatively free of aeolianmaterial (Fig. 9a,b,f).Typical spectra of this region have a well defined pyroxene signature,as visible in Fig. 3. In contrast, the high albedo region in the surround-


ings consists of low inertiamaterial that appears featureless in OMEGAspectra. It most likely corresponds to rocks mantled by eolian depositssuch as bright dust. Thewindowof lowalbedo rocks therefore signals alocation where the spectral composition can be explored. The highalbedo (0.19) compared to the Noctis plains (below 0.15) could sug-gest that a thin dust mantle (b100 µm thick) is present, but not insufficient proportion to mask the detection of pyroxene.

The overall topography of the Echus Chasma floor is flat, sug-gesting that a low viscosity fluid filled the trough (Fig. 9c). At the scaleof MOC images, surface morphology is complex and includes pressureridges, straight fractures, and rafted plates (Fig. 9a,b). Individualplates are several tens of meters to several kilometers wide, relativelysmooth, and separated by rougher and brighter terrains. The smoothplates often seem to fit together like a jigsaw puzzle. Therefore, theseplates can be interpreted to be pieces of a ruptured crust that haverafted over a fluid medium. Pressure ridges are interpreted to exist atthe edges of plates, since these plates are rammed into each otherduring emplacement. These characteristics are typical of platy-ridgedflows, as described by Keszthelyi et al. (2000). In summary, this plaindisplays mineralogical features and morphologic patterns thatcorresponds to a volcanic filling. The rough texture observed onMOC images, the lack of sand dunes, and the relatively high thermalinertia favor a rocky bedrock locally exposed without a thick dustmantle. The area of the plains defined by the OMEGA signature isapproximately 2000 km2.

5. Determination of ages and compositions

5.1. Ages

Ages are obtained counting craters in the three regions studiedaccording to the diagram and methods described in Hartmann andNeukum (2001). This diagram divides the distribution range (craterdiameter) into log intervals incremented in square root 2. Using studiesfrom the Moon, meteoroid distributions and impact frequency,predicted “isochrons,” or crater size-frequency distributions, havebeen derived for well-preserved surfaces of various ages, such as 1 Ga,



Fig. 8. (a) An HRSC close-up (#1977) on the southern part of the smooth plain NL2.Notice the disruption of the surface in the center relative to the surrounding flat area.(b) MOC image #R0800697 of this disrupted area. Three plates N500 mwide are visiblewith their smooth texture. They are separated by rougher terrains. (c) MOC image#R0601602 of the smooth plain. R represents a sinuous ridge. The terrain in the centerand in the left part of the image displays fine texture with flow patterns.


ARTICLE IN PRESS


100 Ma, 10 Ma, and so on. Determined ages are valid for fresh surfacesnot affected by erosion or deposition. Crater counts should followisochrons when the surface is fresh, while crater counts crossingisochrons indicate a surface that has been subsequently modified.Countswere done in aGIS software using sinusoidal projections of HRSCandMOC images formeasuring crater diameters and the areaof theunit.

NL1 crater counts are determined for over 250 km2 on the easternpart of theplain (Fig. 5).We limit counts toHRSC data due to the paucityof images with better resolution in this case. An age followingapproximately the 100 My isochron can be retrieved graphically. Amodel age of 77±21 My is determined using the more precise meansquare rootmethod proposed by Vaucher et al. (2009). The younger agewith this method is due to the fact that smaller craters are more nu-merous, therefore statisticallymore reliable in themeansquaremethod.

NL2 crater counts are determined over an area of 800 km2 fromHRSC images, and over an area of 210 km2 from the threeMOC imagescrossing this part of the plain. The plot reveals a good consistencybetween craters counted in HRSC and MOC images (Fig. 10, circles).The isochron for this plot corresponds to an age of 50 to 100 My. Thisage is well constrained from the large range of craters used. A precisemodel age of 66±5 My is determined using the method of Vaucheret al., 2009.

Echus Chasma Floor crater counts were derived from the relevantHRSC mosaic over an area of 1600 km2. The crater counts are close tothe 100 My isochron (Fig. 10, triangles). The slight shift to a youngerage for smaller craters could be related to the statistically low numberof large craters, or to erosion/deposition processes that influencedpartof the canyon floor, thus removing some small craters. Despite thisslight difference in the isochron slope, ages in the range 50–100 Myprovide reasonable fits. A model age of 52±8 My is determined usingthe mean root squares method (Vaucher et al., 2009), assuming thesmaller bin is not affected by more recent obliteration. This age isconsistent with that of about 70 My established using cumulativemethods (Chapman et al., 2007, submitted for publication).

Plots of crater density follow isochrons well enough to confirmthat the ages correspond to the formation of the plains, or the plotswould have crossed isochrons without following them. One shouldremember that the absolute ages have uncertainties of a factor 3 to 4(e.g. Hartmann and Neukum, 2001). In addition, the correction factordue to a decreasing impact flux through time might underestimatethese ages by a factor of roughly 3 (Quantin et al., 2007). Even withthese limitations, the three studied plains are still extremely youngrelative to the majority of the Martian surface (300 My is an age wellinto in the Late Amazonian epoch). The estimated ages can becompared to other studies of volcanic centers on Mars. These ages areamong the youngest found on Mars for volcanic plains, and are of thesame order as those proposed for the calderas of Arsia Mons orOlympus Mons in the Tharsis region (e.g., Neukum et al., 2004) andvolcanic plains in Central Elysium Planitia (Vaucher et al., 2009).

5.2. Composition

The mineralogy derived from spectra is shown in Table 1.Plagioclases abundance, inferred by OMEGA is included in a mineralgroup referred to as “neutral components,” for which the abundanceis the combined abundance of spectrally featureless phases in the nearinfrared (plagioclases, high silica phases and quartz). Plagioclases areexpected to be predominant in this group for smooth lava plains.

Grain sizes were determined in models from a few tens to a fewhundreds of micrometers, except for dust which was below 10 μm.Results indicate that plagioclase, HCP and LCP are the dominantminerals for the three regions listed in Table 1. A contribution of dustis required in two regions. Olivine was not required in the fit ofspectra. This shows that olivinewas below the detection limit (~5% forgrain size of 100 µm, or a slightly higher proportion for grains below10 µm). The derived abundances of plagioclases and pyroxenes (40 to



Fig. 9. (a) MOC R0100171 of the Chasma floor. (b) MOC #E1701699 of the Chasma floor. Both close-ups show ridges and plates suggesting a solidified crust floating over a fluidmedium. (c) The THEMIS daytime mosaic with MOLA contours superimposed (250 m intervals). (d) The THEMIS daytime mosaic with the THEMIS nighttime mosaic superimposedin color showing warmer locations at night. (e) An OMEGA pyroxene map superimposed on the THEMIS daytimemosaic. (f) An HRSCmosaic (orbit #71 and #5379) of Echus Chasmafloor. Mars Express orbits used for OMEGA data processing are #581 and #1257.


ARTICLE IN PRESS

60% plagioclases, 30 to 40% pyroxenes) are typical of basalticcomposition. The three regions also display a very low LCP/HCPratio, some of the lowest ratios known to date on Mars. Indeed, most


Noachian crustal rocks or Hesperian aged plains have LCP/HCP ratiosof 1:1 to 1:4 (Poulet et al., 2009b), whereas NL1 reaches a ratio of 1:6(17% in Table 1). Such characteristics were interpreted to support the



Fig. 10. Crater counts for the three studied plains presented on Hartmann isochrons. “n”is the total number of craters counted for each image. Error bars (1σ) only includedwhen larger than circles, triangles and squares. Bin divisions in the horizontal axisinclude 500 m, 707 m, 1 km, 1.4 km, 2 km, 2.8 km, and so on. The vertical axis gives thedensity of craters (number of craters/km2 in each bin).


ARTICLE IN PRESS

idea that there is a change in the LCP/HCP ratio through Martiangeological history, with an enrichment in HCP in recent lava flows(Poulet et al., 2009b).

6. Discussion

6.1. The origin of recent volcanism inside the Tharsis area

Recent volcanism in the Tharsis region is generally considered tobe limited to the main volcanic edifices, such as Arsia or OlympusMons and surrounding plains. In our study, we have found two pre-viously undescribed plains b100 My old. These plains were previouslymapped as undifferentiated floor material of Early to MiddleAmazonian age and of diverse nature (Scott and Tanaka, 1986). Norecent volcanism has ever been reported inside the Valles Marineris–Noctis Labyrinthus troughs, despite the large amount of Hesperian agevolcanism surrounding these areas (e.g., Baptista et al., 2008). Inaddition, Echus Chasma floor has been confirmed to be covered byvolcanic terrains of similar ages. The location of the three studiedregions, much further to the East than the main volcanic centers ofTharsis, is a feature of interest in itself to understand the evolution ofValles Marineris Chasmata.

The Echus Chasma floor could have been filled from lava flowscoming from thewesternplateauof the Echus-Kasei systemof troughs,as suggested by Viking based mapping (Scott and Tanaka, 1986). Suchan observationwould suggest that the volcanic centers are actually onthe Tharsis plateau, close to the volcanoes. Nevertheless, fractures andlandforms observed on the bottom of Echus Chasma north of thestudied area suggest that lavaswere also extruded from fissures on thecanyon bottom (see Chapman et al., 2007, submitted for publicationfor more details), not only from the western plateaus. This suggeststhat local magmatic sources also exist below the canyon floor. The twoChasmata of Noctis Labyrinthuswere not filled from the plateaus sincethere is no sign of lavas flowing into the canyons. NL1 displays volcanic


vents, thus showing that the origin of the volcanism is local, below thecanyon floor.

As the origin of the volcanism is not the focus of our study, we donot enter into in-depth discussions on that question, but the observedvolcanic activity, well to the east of the Tharsis area, should be takeninto account in geophysical models attempting to explain recentvolcanism on Mars (e.g., Schumacher and Breuer, 2007). Possibleexplanations include the strong faulting in this region, especially inNoctis Labyrinthus, compared to regions devoid of widespreadfaulting, and, the thinner crust on the floor of Chasmata than on theplateau (N5 km of difference of elevation). Both of these features mayenable magmatic activity to reach the surface.

6.2. The composition of platy-ridged flows

The existence of platy-ridged flows with rafted slabs and pressureridges is well known on Mars (i.e., Keszthelyi et al., 2000). However,this type of pattern is not frequent for terrestrial volcanism, and hasbeen discussed almost exclusively for flows in Iceland (Keszthelyiet al., 2004, Haack et al., 2006). Therefore, not all workers agree thatthese features onMars are lava flows.Mudflows and pack ice (icebergsfloating over water) have a very similar shape, with rafted slabs of iceover water or rocks over mud. Such a scenario has been proposed forplaty flows seen in Kasei Vallis by Woodworth-Lynas and Guigné(2003) and Williams and Malin (2004). In this respect we note thatKasei Vallis originates in Echus Chasma where our platy flows areobserved. In addition, Central Elysium Planitia plains have beenproposed to have been filled by a frozen sea displaying rafted plates(Murray et al., 2005). The problem of these contradictory interpreta-tions is that they are based on geomorphic landforms of very similarshape. No spectral study of platy-ridged flows has been performed todiscriminate these hypotheses, principally because all of these regionsare blanketed by dust. However, in the regions studied here, weobserve two plains that display platy-ridged textures: the EasternNoctis Chasma (NL2) and Echus Chasma (Figs. 8 and 9 respectively).Both of these areas display a mineralogy of basaltic lava flows withpredominant plagioclases and pyroxenes. While we cannot confirmthat all platy-ridge flows are magmatic in origin, our data provide thefirst unambiguous evidence that suchmorphological features onMarsmay indeed be associated with basaltic lava flows.

Models of platy-ridged lava flows generally show a viscosity in therange of 100–1000 Pa s. (Keszthelyi et al., 2000). Viscosities of below103 Pa s, and yield strengths less than 200 Pa are found in the CentralElysium Planitia (Vaucher et al., 2009), a location where extensiveplaty flows have also been reported (Keszthelyi et al., 2000).Unusually low viscosity or low yield strengths for silicate magmamay be related to different parameters, including a high temperatureor a low proportion of Si. The mean compositions of the two plains arethat of classical basalts, and not that of an ultrabasic rock. Althoughthe mineralogy of platy-ridged flows has been established at thesurface of the flows, this does not give themineralogy at the bottom offlows. A downward segregation of olivine may have taken place, thusbeing a possible explanation for the lack of olivine detection in theOMEGA spectra. However, the fact that olivine is systematicallyabsent, or at least in limited proportion in small grains, even in erodedparts of NL2, and that the detection of pyroxene, another densemineral, is clear, are both arguments that plead against a loss ofolivine from the surface. Thus, the lack of detection of olivine suggeststhat an ultrabasic composition was not necessary to explain the lavasfluidity. A similar conclusion has been drawn from Icelandic exampleswhere platy flows correspond to tholeiitic basalts (Keszthelyi et al.,2004). Thus, mechanisms such as deep crustal melting, dissolvedvolatiles (Vaucher et al., 2009), or high effusion rates (Keszthelyi et al.,2000) are necessary to explain the fluidity of platy-ridged flows ratherthan an ultramafic composition.




ARTICLE IN PRESS

6.3. The composition of young lava flows compared to SNC meteorites

Shergottite meteorites generally accepted to be of Martian originare commonly dated at ages between 175 and 475 Ma (e.g., Nyquistet al., 2001), a range comparable to the youngest volcanic areas onMars. However, the link between basaltic shergottites, those expectedto outcrop near the surface, and Martian volcanism as observed fromorbit has not been very successful, especially due to the extensive dustmantle overlying young volcanic regions (e.g., Hamilton et al., 2003).Since the mineralogy of rocks from older regions does not fit themineralogy of SNC meteorites, a conclusion has emerged thatshergottite compositions are not found in spectral data because theyoungest lava flows cannot be analyzed (Hamilton et al., 2003). In ourstudy, we identified and determined the mineralogy of three areas ofnonmantled lava flows from the Late Amazonian, with ages b100 My.Determined ages are slightly younger than the youngest ages usuallyconsidered for shergottites, but these ages are not significantlydifferent due to the uncertainties associated with the determinationof absolute ages using the crater count method. Our results indicatethat the mineralogy of these three lava plains studied is alsoinconsistent with the mineralogy of basaltic shergottites in terms ofLCP/HCP ratios. Indeed, most of the basaltic shergottites are richer inLCP when compared to the proportions derived from the relevantOMEGA data. For example, data for Shergottites suggest LCP/HCPratios from 1:1 to 18:1 (McSween and Jarosewich, 1983; Barrat et al.,2002; Taylor et al., 2002, Hamilton et al., 2003). In contrast, ourstudied areas have average LCP/HCP ratios from 1:4 to 1:6. Otherregions of recent volcanic activity may have different ratios closer tothose observed in SNCs. Nevertheless, shergottites ages of approxi-mately 4.0 Gy determined by Bouvier et al. (2005) may be analternative explanation. If this is the case, the observed discrepancy isexplained because shergottites would not be relevant for a compar-ison with recent volcanism.

7. Conclusion

We analyzed the morphology and mineralogy of some of the mostrecent volcanic areas in the East Tharsis region of Mars. Findingsinclude the following:

(i) Previously unknown volcanic plains 50 to 100My old are foundin two canyon floors of Noctis Labyrinthus, while similar recentages were confirmed for the floor of Echus Chasma.

(ii) These volcanic plains are located N1000 km east of previouslyknown Late Amazonian volcanic centers of the Tharsis region,an observation to include in models of recent Mars volcanism.

(iii) The platy-ridged textures observed in two of the studiedregions are associated with volcanic lava flows as seen from thebasaltic mineralogy coupled with volcanic landforms.

(iv) The apparent low viscosity of platy lava flows may be due toparameters other than low Si content of ultramafic rocksbecause of the limited olivine content (b5% assuming ~100 µmgrains) and the classical basaltic mineralogy obtained.

(v) The low-calcium/high calcium pyroxene ratios found bymodeling for these young lava flows are different from thatof putative young Martian meteorites (basaltic shergottites).

Finally, our study shows that in the absence of a widespread recentregion devoid of dust, we need to focus on small dust-free windows tocharacterize the mineralogy of recent volcanism on Mars.

References

Ansan, V., Mangold, N., Masson, Ph., Neukum, G., 2008. Topography of valley networkson Mars fromMars Express high resolution stereo camera digital elevation models.J. Geophys. Res. 113, E07006. doi:10.1029/2007JE002986.


Baptista, A.R., Mangold, N., Ansan, V., Baratoux, D., Lognonné, P., Alves, E.I., Williams, D.A.,Bleacher, J.E.,Masson, P., Neukum,G., 2008. A swarmof small shield volcanoesonSyriaPlanum, Mars. J. Geophys. Res. 113, E09010. doi:10.1029/2007JE002945.

Barrat, J.A., Jambon, A., Bohn, M., Gillet, P., Sautter, V., Gopel, C., Lesourd, M., Keller, F.,2002. Petrology and chemistry of the pricritic shergottite NorthWest Africa 1068(NWA 1068). Geochim. Cosmochim. Acta 66, 3505–3518.

Berman, D.C., Hartmann, W.K., 2002. Recent fluvial, volcanic, and tectonic activity onthe cerberus plains of Mars. Icarus 159, 1–17.

Bibring, J.-P., et al., 2004. OMEGA: Observatoire pour la Minéralogie, l'Eau, les Glaces etl'Activité. In: Wilson, A. (Ed.), Mars Express: The Scientific Payload: Eur. SpaceAgency Spec. Publ., ESA-SP, vol. 1240, pp. 37–49.

Bouvier, A., Blichert-Toft, J., Vervoort, J.D., Albarède, 2005. The age of SNC meteoritesand the antiquity of the Martian surface. Earth Planet. Sci. Lett. 240, 221–233.

Chapman, M.G., Neukum, G., Werner, S.C., van Gasselt, S., Dumke, A., Zuschneid, W.,Michael, G., 2007. Echus Chasma and Kasei Valles, Mars: new data and geologicinterpretations. Lunar Planet. Sci. Conf. 38th, abst. 1407.

Chapman,M.G., Neukum, G., Dumke, A., Michaels, G., van Gasselt, S., Kneissl, T., Zuschneid,W., andHauber, E., submitted for publication,Amazonian geologic history of theEchusChasma and Kasei Valles system onMars: new data and interpretations, part 2, EarthPlanet. Sci. Let.

Christensen, P.R., et al., 2003. Morphology and composition of the surface of Mars: MarsOdysseyTHEMIS results. Science300(5628), 2056–2061. doi:10.1126/science.1080885.

Greeley, R., Spudis, P., 1981. Volcanism on Mars. Rev. Geophys. 19 (1), 13–41.Gwinner, K.F., Scholten, Gise, B., Oberst, J., Jaumann, R., Spiegel, R., Schmidt, R., Neukum, G.,

2005. Hochauflösende Digitale Geländemodelle der Marsoberfläche auf der Grund-lagevonMarsExpressHRSCdaten, Photogrammetrie FernerkundungGeoinformation(PFG) 5/2005, S387–394, Germany.

Haack, H., Rossi, M.J., Dall, J., 2006. SAR mapping of Burfellshraun: a terrestrial analog forrecent volcanism on Mars. J. Geophys. Res. 111, E06S13. doi:10.1029/2005JE002536.

Hamilton, V.E., Christensen, P.R., McSween Jr., H.Y., Bandfield, J.L., 2003. Searching forthe source regions of Martian meteorites using MGS TES: integrating Martianmeteorites into the global distribution of igneous materials on Mars. Meteorit.Planet. Sci. 38, 871–886.

Hartmann, W.K., Neukum, G., 2001. Cratering chronology and evolution of Mars. SpaceSci. Rev. 96 (1–4), 165–194.

Hodges, C.A., Moore, H.J., 1994. Atlas of volcanic landforms onMars. USGS Prof. Paper,p. 1534. Washington.

Keszthelyi, L., McEwen, A., Thordarson, T., 2000. Terrestrial analogs and thermal modelsfor Martian flood lavas. J. Geophys. Res. 105 (E6), 15,027–15,049.

Keszthelyi, L., Thordarson, T., McEwen, A., Haack, H., Guilbaud, M.N., Self, S., Rossi, M.J.,2004. Iceland analogs to martian flood lavas. Geochim. Geophys. Geosyst. 5 (11).doi:10.1029/2004GC000758, 32 pp.

Lanagan, P.D., McEwen, A.S., Keszthelyi, L.P., Thordarson, T., 2001. Rootless cones onMarsindicating thepresenceof shallowequatorial ground ice in recent times. Geophys. Res.Lett. 28, 2365–2368.

Malin, M.C., Danielson, G.E., Ingersoll, A.P., Masursky, H., Veverka, J., Ravine, M.A.,Soulanille, T.A., 1992. Mars observer camera. J. Geophys. Res. 97 (E5), 7699–7718.

Mangold, N., Gendrin, A., Gondet, B., Le Mouelic, S., Quantin, C., Ansan, V., Bibring, J.P.,Langevin, Y., Masson, P., Neukum, G., 2008. Spectral and geological study of thesulfate rich region of West Candor Chasma, Mars. Icarus 194 (2), 519–543.

McSween, H.Y., Jarosewich, E., 1983. Petrogenesis of the Elephant Moraine A79001meteorite multiple magma pulses on the shergottite parent body. Geochim. Cosmo-chim. Acta 47, 1501–1513.

Meresse, S., Costard, F.,Mangold,N.,Masson, P., Neukum,G., 2008. Formation andevolutionof the chaotic terrains by subsidence and magmatism: Hydraotes Chaos, Mars. Icarus194 (2), 487–500.

Murray, J.B.,Muller, J.P., Neukum,G.,Werner, S., VanGasselt, S., Hauber, E.,Markiewicz,W.,Head, J.W., Foing, B., Page, D., Mitchell, K.L., Portyankina, G., et al., 2005. Evidence fromtheMars Express high resolution stereo camera for a frozen sea close toMars' equator.Nature 434 (7031), 352–356.

Mustard, J.F., Poulet, F., Gendrin, A., Bibring, J.-P., Langevin, Y., Gondet, B., Mangold, N.,Bellucci, G., Altieri, F., 2005.Olivine andpyroxenediversity in the crust ofMars. Science307 (5715), 1594–1597. doi:10.1126/science.1109098.

Neukum, G., Jaumann, R., 2004. HRSC: the high resolution stereo camera of Mars Express.In: Wilson, A. (Ed.), Mars Express: The Scientific Payload: Eur. Space Agency Spec.Publ., ESA-SP, vol. 1240, pp. 17–35.

Neukum, G., et al., 2004. Recent and episodic volcanic and glacial activity onMars revealedby the high resolution stereo camera. Nature 432 (7020), 719–729.

Nyquist, L.E., Bogard, D.D., Shih, C.-Y., Greshake, A., Stöffler, D., Eugster, O., 2001. Agesand geologic histories of Martian meteorites. Space Sci. Rev. 96, 105–164.

Poulet, F., Erard, S., 2004. Nonlinear spectral mixing: quantitative analysis of laboratorymineral mixtures. J. Geophys. Res. 109 (E2). doi:10.1029/2003JE002179.

Poulet, F., Mangold, N., Erard, S., 2003. A new view of dark Martian regions fromgeomorphic and spectrocopic analysis of Syrtis Major. Astron. Astrophys. 412,L19–L23. doi:10.1051/0004-6361:20031661.

Poulet, F., Gomez, C., Bibring, J.-P., Langevin, Y., Gondet, B., Pinet, P., Belluci, G., Mustard, J.,2007.Martian surfacemineralogy fromOMEGAonboard theMars Express spacecraft:global mineral maps. J. Geophys. Res. 112, E08S02. doi:10.1029/2006JE002840.

Poulet, F., Bibring, J.-P, Langevin, Y.,Mustard, J.F.,Mangold,N., Vincendon,M.,Gondet, B., Pinet,P., Bardintzeff, J.-M., Platevoet, B., 2009a. Quantitative compositional analysis of Martianmafic regions using MEx/OMEGA reflectance data: 1. Methodology, uncertainties andexamples of application. Icarus 69–83. doi:10.1016/j.icarus.2008.12.025.

Poulet, F.,Mangold,N., Platevoet, B., Bardintzeff, J.-M., Sautter,V.,Mustard, J.F., Bibring, J.-P.,Pinet, P., Langevin, Y., Gondet, B., Aléon-Toppani, A., 2009b. Quantitative composi-tional analysis of Martian mafic regions using the MEx/OMEGA reflectance data: 2.Petrological implications. Icarus 84–101. doi:10.1016/j.icarus.2008.12.042.



http://dx.doi.org/10.1029/2007JE002945

http://dx.doi.org/10.1126/science.1080885

http://dx.doi.org/10.1029/2005JE002536

http://dx.doi.org/10.1029/2004GC000758

http://dx.doi.org/10.1126/science.1109098

http://dx.doi.org/10.1029/2003JE002179

http://dx.doi.org/10.1051/0004-6361:20031661

http://dx.doi.org/10.1029/2006JE002840

http://dx.doi.org/10.1016/j.icarus.2008.12.025


http://dx.doi.org/10.1029/2007JE002986


ARTICLE IN PRESS

Putzig, N.E., Mellon, M.T., Kretke, K.A., Arvidson, R.E., 2005. Global thermal inertia andsurface properties ofMars from theMGSmappingmission. Icarus 173 (2), 325–341.doi:10.1016/j.icarus.2004.08.017.

Quantin, C., Mangold, N., Hartmann,W.K., Allemand, P., 2007. Possible long-termdeclinein impact rates. 1. Martian geological data. Icarus 186 (1), 1–10.

Scholten, F., Gwinner, K., Roasch, T., Matz, K.D., Wählisch, M., Giese, B., Oberst, J., Jaumann,R., Neukum, G., 2005. Mars Express HRSC data processing. Photogramm. Eng. RemoteSensing 71, 1143–1152.

Schumacher, S., Breuer, D., 2007. An alternative mechanism for recent volcanism onMars. Geophys. Res. Lett. 34, L14202.

Scott, D.H., Tanaka, K.L., 1986. Geologic map of western equatorial region of Mars. U.S.Geol. Surv. Misc. Invest Ser., Map I-1802-A.

Seidelmann, P.K., Abablakin, V.K., Bursa, M., Davies, M.E., De Bergh, C., Leiske, J.H., Oberst, J.,Simon, J.L., Standish, E.M., Stooke, P., Thomas, P.C., 2002. Report of the IAU/IAGworkinggroupon cartographic coordinates and rotational elements of theplanets andsatellites:2000. Celest. Mech. Dyn. Astron. 82, 83–110. doi:10.1023/A:1013939327465.


Smith, D.E., et al., 1999. The global topography of Mars and implications for surfaceevolution. Science 284, 1495–1503. doi:10.1126/science.284.5419.1495.

Stockstill-Cahill, K.R., Scott, A.F., Hamilton, V.E., 2008. A study of low-albedo depositswithin Amazonis Planitia craters: evidence for locally derived ultramafic to maficmaterials. J. Geophys. Res. 113, E07008. doi:10.1029/2007JE003036.

Taylor, L.A., et al., 2002. Martianmeteorite Dhofar 9: a new shergottite. Meteorit. Planet.Sci. 37, 1107–1128.

Vaucher, J.D., Baratoux, Toplis, M.J., Pinet, P., Mangold, N., Kurita, K., 2009. The mor-phologies of volcanic landforms at Central Elysium Planitia: evidence for recent andfluid lavas on Mars. Icarus 200, 39–51. doi:10.1016/j.icarus.2008.11.005.

Werner, S.C., 2009. The global martian volcanic evolutionary history. Icarus 201, 44–68.Williams, R.M.E., Malin, M.C., 2004. Evidence for late-stage fluvial activity in Kasei

Vallis, Mars. J. Geophys. Res. 109, E06001. doi:10.1029/2003JE002178.Woodworth-Lynas, C., Guigné, Y., 2003. Ice keel scour marks on Mars: evidence for

floating and grounding ice flows in Kasei Vallis. Oceanography 16, 90–97.




http://dx.doi.org/10.1023/A:1013939327465

http://dx.doi.org/10.1029/2003JE002178


http://dx.doi.org/10.1029/2007JE003036

http://dx.doi.org/10.1126/science.284.5419.1495

Documents

Mineralogy of recent volcanic plains in the Tharsis region, Mars, and implications for platy-ridged flow composition