Planetary and Space Science · 18 m/pixel in the full resolution targeted (FRT) mode and at ∼40 m/pixel in the half-resolution short or long (HRS/HRL) mode (Murchie et al., 2009c)

Planetary and Space Science 86 (2013) 130–149

Contents lists available at ScienceDirect

Planetary and Space Science

0032-06http://d

n CorrAve, Mo

E-m

journal homepage: www.elsevier.com/locate/pss

What the ancient phyllosilicates at Mawrth Vallis can tell us aboutpossible habitability on early Mars

Janice L. Bishop a,b,n, Damien Loizeau c, Nancy K. McKeown d, Lee Saper a,e, M. Darby Dyar f,David J. Des Marais b, Mario Parente g, Scott L. Murchie h

a Carl Sagan Center, The SETI Institute, Mountain View, CA 94043, USAb Exobiology Branch, NASA-Ames Research Center, Moffett Field, CA 94035, USAc ESA-ESTEC, Noordwijk, The Netherlandsd Department of Physical Sciences, Grant MacEwan University, Edmonton, AB, Canada T5J 4S2e Department of Geological Sciences, Brown University, Providence, RI 02916, USAf Department of Astronomy, Mount Holyoke College, 50 College Street, South Hadley, MA 01075, USAg Department of Electrical and Computer Engineering, University of Massachusetts, Amherst, MA 01003, USAh Applied Physics Laboratory, Laurel, MD 20723, USA

a r t i c l e i n f o

Article history:Received 26 March 2012Received in revised form6 September 2012Accepted 15 May 2013Available online 24 May 2013

Keywords:MarsPhyllosilicatesReflectance spectroscopyWaterHabitability

33/$ - see front matter & 2013 Elsevier Ltd. Ax.doi.org/10.1016/j.pss.2013.05.006

esponding author at: The SETI Institute, Carluntain View, CA 94043, United States. Tel.: +ail address: [email protected] (J.L. Bishop).

a b s t r a c t

Phyllosilicate deposits on Mars provide an opportunity to evaluate aqueous activity and the possibilitythat habitable environments may have existed during the Noachian period there. Analysis of hyper-spectral visible/near-infrared (VNIR) Mars Reconnaissance Orbiter (MRO) Compact ReconnaissanceImaging Spectrometer for Mars (CRISM) images has shown thick, complex profiles of phyllosilicates atMawrth Vallis, Mars that are consistent with long-term aqueous activity and active chemistry. Theancient phyllosilicates in places such as this could have served as reaction centers for organic molecules.Previous experiments even suggest that phyllosilicates could have played a role in the origin of life.Regardless of whether life formed on early Mars or not, evaluating the type and thickness of clay-bearingunits on Mars provides insights into plausible aqueous processes and chemical conditions both duringthe time of formation of the phyllosilicates, but also the subsequent period following their formation.The phyllosilicate outcrops at Mawrth Vallis extend across a broad (∼1000 km) region and exhibit aconsistent general trend of Al-phyllosilicates and amorphous Al/Si species at the top of the clay profileand Fe/Mg-phyllosilicates on the bottom. This implies either a change in water chemistry, a change inmaterial being altered, or an alteration profile where the upper clays were leached and altered moresignificantly than those below. A change in iron in the phyllosilicate units is also observed such that anFe2+-bearing unit is frequently observed between the Fe3+- and Mg-rich phyllosilicates below and theAl/Si-rich materials above. Abrupt changes in chemistry like this are often indicative of biogeochemicalactivity on Earth. Possible microbe-clay interactions are considered in comparison with the CRISMobservations. This study evaluates CRISM spectra from four images of different outcrops across theMawrth Vallis region and evaluates the observed phyllosilicates and clay components in terms ofplausible aqueous and microbial processes and the potential for retention of biosignatures, if present.

& 2013 Elsevier Ltd. All rights reserved.

1. Introduction

Analyses of Mars Reconnaissance Orbiter (MRO) CompactReconnaissance Imaging Spectrometer for Mars (CRISM) data haverevealed a wide range of aqueous processes taking place on Mars,largely during the Noachian and Hesperian period (Murchie et al.,2009b). The Mawrth Vallis region harbors some of the largest

ll rights reserved.

Sagan Center, 189 Bernardo1 65 08100222.

phyllosilicate outcrops on Mars with abundant exposures of manyaqueous minerals and phases (e.g. Poulet et al., 2005; Loizeauet al., 2007; Bishop et al., 2008b; McKeown et al., 2009; Michalskiet al., 2010; Noe Dobrea et al., 2010). The Mawrth Vallis region issituated at the boundary of the southern highlands and thenorthern lowlands near 22–251N and 17–211W and the elevationvaries from approximately −2700 to −3600 m (e.g. Loizeau et al.,2007). The region was cut by the Mawrth Vallis channel and the∼100-km wide Oyama crater. Extensive phyllosilicate outcropshave been identified at Mawrth Vallis in Mars Express OMEGA(Observatoire pour la Minéralogie, L'Eau, les Glaces et l'Activité)and MRO CRISM images using spectral absorptions at 1.38–1.42,

www.sciencedirect.com/science/journal/00320633

www.elsevier.com/locate/pss

http://dx.doi.org/10.1016/j.pss.2013.05.006



http://crossmark.crossref.org/dialog/?doi=10.1016/j.pss.2013.05.006&domain=pdf



mailto:[email protected]


J.L. Bishop et al. / Planetary and Space Science 86 (2013) 130–149 131

1.91–1.93, 2.16–2.33 and 2.39–2.41 mm. The Mawrth Vallis regioncontains large exposures of Fe/Mg-smectite as the deepest phyllo-silicate unit within the ancient cratered terrain. This exists as athick, pervasive unit that was subsequently leached to or coveredby material rich in Al-phyllosilicates and hydrated silica (Bishopet al., 2008b; Wray et al., 2008; McKeown et al., 2009; Loizeauet al., 2010; Michalski et al., 2010).

This study builds off the earlier ones by evaluating mosaicsof CRISM hyperspectral visible/near-infrared (VNIR, 0.4–4 mm)images in order to gain an understanding of broader mineralogicaltrends in two large regions of Mawrth Vallis. The recently availableTRR3 calibration of the images presented here has enabled moreaccurate identification of the metal-OH stretch plus bend combi-nation features in the range 2.1–2.5 mm. We use these improveddata especially for characterization of the Al/Si-rich upper clay unitthat contains a mixture of components. This study also providesinformation about the geochemistry of the aqueous environmentsduring clay formation and subsequent geologic activity, as well asassessment of the potential for biosignature preservation in thedifferent clay-bearing units at Mawrth Vallis. Phyllosilicates, espe-cially dioctahedral smectites, have been the focus of numerouspast studies involving early organic chemistry on the Earth(Pinnavaia and Mortland, 1986; Ferris et al., 1988; Franchi et al.,2003; Ferris, 2005, 2006) and the origin of life (Lawless, 1986). Wepresent the phyllosilicate minerals observed at Mawrth Vallis incontext with these studies in order to understand the potential fororganic chemistry and habitability at Mawrth Vallis during earlyMartian history.

2. Methods

CRISM collects ∼10 km wide images from 0.4 to 3.9 mm at18 m/pixel in the full resolution targeted (FRT) mode and at∼40 m/pixel in the half-resolution short or long (HRS/HRL) mode(Murchie et al., 2009c). CRISM TRR3 images (calibration level 3)were used for this study that include a more robust flat fieldcorrection (Seelos, 2011). Calibration version 3 also corrects for theeffects of unremoved water vapor in observations of the groundcalibration source to which inflight radiometric calibration is

Table 1List of laboratory spectra of minerals used in Figs. 6–7.

Mineral Type ID

Fe/Al-smectite Phyllosilicate JB170Nontronite Phyllosilicate JB175Fe/Mg-smectite Phyllosilicate JB761Hectorite Phyllosilicate JB768Saponite Phyllosilicate USGSLizardite Phyllosilicate JB526

C53Montmorillonite Phyllosilicate JB171Beidellite Phyllosilicate JB919

WG#34Kaolinite Phyllosilicate JB225Halloysite Phyllosilicate JB1049Celadonite Mica JB727Chamosite Chlorite JB739

21DGlauconite Mica JB731Acid-treated smectite (SWa-1 in 6 M HCL 1.2 h)Opal-A JB629Opal-CT opal JB874Allophane JB1021Gibbsite Hydroxide JB1076Jarosite Sulfate JB406Siderite Carbonate JB287Fo70 Olivine JB945

traced, and the effects of slight observation-to-observation irre-producibility in the viewing geometry of the inflight calibrationreference, a small integrating sphere viewed by an internal mirror.In addition calibration version 3 applies a kernel filter that leveragesthe hyperspectral character of the data and the high degree ofredundancy of adjacent spectra to identify stochastic noise in astatistically robust way and interpolate over it. Both the short (S)0.4–1 mm and long (L) 1–3.9 mm wavelength image pairs werecalibrated and evaluated for this study.

The images were processed using the CRISM Analysis Tool (CAT)for ENVI following standard procedures (Murchie et al., 2007,2009c). The data were converted to I/F and variations in illumina-tion were corrected by dividing the I/F image by the cosine ofthe incidence angle (derived from MOLA gridded topography at128 pixels/deg). Atmospheric molecular opacity effects were mini-mized in the L images by dividing by a scaled atmospherictransmission spectrum over Olympus Mons. A denoising algorithmwas applied to the L images that removes vertical stripes by low-pass filtering in the spatial and spectral domains (Parente, 2008).This denoising algorithm optimizes the filtering in order to preservespatial information while cleaning the image. The process thendetects and removes spectral spikes by comparing spectral channelvalues with thresholds and substituting the detected spikes withthe average of the spectral values of the adjacent channels (Parente,2008). Spectral parameters were applied to both the S and L imagesin order to highlight changes in surface composition (Pelkey et al.,2007; Murchie et al., 2009c). Spectra were collected of 5�5 or10�10 pixel clusters and ratioed to spectrally neutral regions in thesame column in order to maximize the spectral contributions fromthe surface minerals.

The images were georeferenced and draped over MOLA terrain inorder to better visualize the relative positions of the phyllosilicate-bearing deposits on the local topography. Several images weremosaicked together across two broad regions of Mawrth Vallis toprovide a regional perspective. 3D surface views of CRISM data werecreated using MOLA data of selected images in order to illustratewhere the spectra were collected.

Laboratory reflectance data of several minerals and relatedmaterials were acquired in previous studies and used here forcomparison with the CRISM data (Table 1). Most of the smectite

Location References

SWa-1, Grant County, WA Bishop et al. (2002a)Sampor, Slovakia Bishop et al. (2002a)Flagstaff Hill, CA Bishop et al. (2008b)SHCa-1, San Bernadino, CA Bishop et al. (2002b)SapCa-1, Ballarat, CA Clark et al. (2007)Cassiar deposit, Bishop et al. (2002b)British Columbia, Canada O'Hanley & Dyar (1993)SAz-1, Arizona Bishop et al. (2008a)Delamar Mine, ID Bishop et al. (2011a)(Glen Silver Pit) Gates (2005)KGa-1 Bishop et al. (2008a)#902474 from S. HillierMojave Desert, CA Bishop et al. (2008a)Ishpeming MI Bishop et al. (2008a)

Hemingway et al. (1984)Hurricane Mountain Bishop et al. (2008a)Prepared in the lab Madejova et al. (2009)Kilauea solfatara Bishop et al. (2005?)Virgin Valley, NV McKeown et al. (2011)Synthesized by L. Baker Bishop et al. (2013)Research grade powder synthesized by Ward'sMineral County, NV Bishop and Murad (2005)Roxbury Iron Mine, Canada Bishop et al. (2008a)San Carlos, AZ

J.L. Bishop et al. / Planetary and Space Science 86 (2013) 130–149132

spectra are from Bishop et al. (2002b, 2002a, 2008a, 2011a). Thesaponite (Mg-smectite) and zeolite (analcime, scolecite and heu-landite) spectra are from the USGS spectral library (Clark et al.,2007). The acid-treated smectite spectrum is from a sample ofSWa-1 (ferruginous smectite) exposed to 6 M HCl for 1.2 h at 80 1C(Madejová et al., 2009). Reflectance spectra of samples from theBishop library were measured at RELAB, Brown University.

3. Results of spectral analyses of CRISM data

Two large phyllosilicate outcrops are observed at Mawrth Vallis(Figs. 1 and 2). Each of these regions contain both Fe/Mg-richphyllosilicates and Al/Si-rich phyllosilicates (and related compo-nents) based on the wavelength of the metal-OH combinationband near 2.2–2.3 mm and the H2O band near 1.9 mm that aredescribed in more detail later. Both types of phyllosilicates

Fig. 1. CRISM over CTX over HRSC for northwestern phyllosilicate outcrop. (a) map oof northwestern phyllosilicate outcrop in the red box and region of eastern phyllosilicG BDI1000VIS, B BD920 parameters resulting in ferric oxide/hydroxide minerals in redG OLINDEX, B BD2210 parameters resulting in the Fe/Mg-smectite unit in orange/red, t(For interpretation of the references to color in this figure legend, the reader is referred

(red/orange and blue units in L images in Figs. 1c and 2b) appearto be correlated well with ferric oxide/hydroxide components (redunit in S image mosaics in Figs. 1b and 2a), while the caprock (darkpurple/black areas in L images in Figs. 1c and 2b) compares wellwith ferrous mineralogy (green in S images in Figs. 1b and 2a). Theregion north of Oyama crater and south of Mawrth Vallis' mouth(Fig. 1) exhibits large exposures of Al-rich phyllosilicates andamorphous hydrated Al/Si components having an M-OH band at∼2.17–2.23 mm. This region lies to the west of the Mawrth Vallischannel. Fe/Mg-smectites are also present at the surface in thisregion and have M–OH bands near 2.3 and 2.4 mm, but are lessexposed than the Al/Si-rich material. The second region is locatedto the east of Oyama crater on the western margin of the MawrthVallis channel (Fig. 2). This region features more widespread Fe/Mg-smectites than Al/Si-rich species on the surface.

Spectral transects across selected images (Figs. 3–5) illustratethe transitions from the Fe/Mg-smectite-rich unit to the Al/Si-rich

f Mawrth Vallis region (credit: Mars Express/HRSC/Google Mars) showing regionate outcrop in the yellow box, (b) CRISM image mosaics prepared with R BD530,and ferrous minerals in green, (c) CRISM image mosaics prepared with R D2300,he ferrous boundary region in green/yellow and the upper Al/Si-rich unit in blue.to the web version of this article.)

Fig. 2. CRISM over CTX over HRSC for eastern phyllosilicate outcrop. (a) CRISM image mosaics prepared with R BD530, G BDI1000VIS, B BD920 parameters resulting in ferricoxide/hydroxide minerals in red and ferrous minerals in green, (b) CRISM image mosaics prepared with R D2300, G OLINDEX, B BD2210 parameters resulting in the Fe/Mg-smectite unit in orange/red, the ferrous boundary region in green/yellow and the upper Al/Si-rich unit in blue. (For interpretation of the references to color in this figurelegend, the reader is referred to the web version of this article.)


unit. These transect spectra are all ratioed to the same denomi-nator spectrum within each image so that any artifacts introducedthrough ratioing are the same in all spectra from that image. 3Dsurface views are shown for pairs of CRISM images. The S imagehighlights ferric oxide/hydroxide minerals in red and ferrousminerals in green, while the L image displays the Fe/Mg-smectiteunit in orange/red, the ferrous boundary region in green/yellowand the upper Al/Si-rich unit in blue.

Shown in Fig. 3 are spectra along the indicated transect from animage in the second region (Fig. 2) where the Fe/Mg-smectitesignatures are particularly strong (spectra 1 and 2). The bands hereare observed near 1.43, 1.92, 2.29, and 2.40 mm and are consistentwith nontronite or Mg-rich nontronite (e.g. Bishop et al., 2008a).Spectra with these features are found in units with differentalbedos. Bordering this unit is a material with weaker Fe/Mg-smectitefeatures (spectrum 3) and a positive slope from ∼1–1.8 mm that ischaracteristic of bands due to electronic excitations in ferrous

materials (e.g. Burns, 1993). Following this is another material with asimilar positive slope and a band near 2.2 mm (spectrum 4). Note thatall of these spectra include some degree of positive slope due to theratioing process, as the denominator spectrum from a spectrallyneutral region has some dust covering it giving rise to a negativeslope. However, the changes in magnitude of slope here indicatechanges in slope in the spectral units as they are all ratioed tothe same denominator spectrum. Additional spectra are observedwith bands at 1.41, 1.92, and a broad feature from 2.17–2.21 mm(spectrum 5). This is consistent with a mixture of Al-phyllosilicatesand amorphous hydrated Al/Si-rich materials including altered silicaglass and opal (e.g. McKeown et al., 2011).

Spectra from CRISM images in the northern phyllosilicate outcrop(Fig. 1) are shown in Figs. 4 and 5. Transects are shown again in orderto illustrate spectral changes due to surface composition along astratigraphic column that are exposed due to erosion. The Fe/Mg-smectite spectra in this region (e.g. images FRT0000AA7D and

Fig. 3. 3D CRISM images and transect spectra of FRT000089F7. The spectra are numbered in order along the transect. (a) CRISM draped over MOLA elevations (5X verticalexageration) prepared with R BD530, G BDI1000VIS, B BD920 parameters, (b) CRISM draped over MOLA elevations prepared with R D2300, G OLINDEX, B BD2210 parameters(white line marks transect where spectra were collected), (c) CRISM I/F spectra extracted from one column of the S image (0.4–1.04 mm) and ratioed CRISM I/F spectra fromthe same column of the L image (1.04–2.65 mm) rescaled to the S image value at 1.04 mm and an insert with ratioed (non-scaled) CRISM/I/F spectra from 1.8–2.6 mm.


FRT0000863E) often exhibit weaker bands, compared to spectra ofthe SE phyllosilicate outcrop (Fig. 2), but these bands are located atsimilar wavelength positions: 1.43, 1.91–1.92, 2.29–2.30, and 2.40–2.41 mm (Fig. 4 spectra 1 and 2, Fig. 5 spectrum 1). The upper Al/Si-rich region contains some spectra with broad bands near 2.2 mm(Fig. 4 spectra 5–7) indicating materials like opal or allophane(hydrated Al/Si protoclay) as well as some regions with narrowerbands more consistent with kaolinite and montmoril-lonite (Fig. 5 spectra 3 and 4). There are some regions in imageFRT0000863E that contain a doublet feature at 2.20 and 2.26 mm(Fig. 5, spectrum 2) indicating the presence of an additional compo-nent (e.g. Fe–OH site) mixed with the Al-phyllosilicates. Changes inslope were also observed along the transect from Fe/Mg-smectite toAl/Si-rich material for images FRT0000AA7D and FRT0000863E ofthe northern phyllosilicate outcrop region. This change in slopeincreases at the border of the two phyllosilicate units and is presentin both the Fe/Mg-smectite type spectra and spectra of the Al/Si-richmaterial.

CRISM spectra of several sites at Mawrth Vallis are shown inFig. 6 compared with laboratory reflectance spectra. Four exam-ples of the Fe/Mg-smectite type spectra are shown in Fig. 6a thatexhibit a band near 2.3 mm due to an Fe/Mg(OH)2 combination

stretching and bending band. Spectra 3 and 4 are more typicalof this unit and contain features consistent with nontronite orMg-bearing nontronite. The OH stretching overtone occurs near1.43 mm, which is consistent with nontronite rather than saponiteor other Mg-rich phyllosilicates (e.g. Bishop et al., 2002a, 2008a).The H2O combination band occurs at 1.92 mm for all four spectra;this band normally occurs at 1.91 mm for smectites and is absent inserpentines or chlorites (e.g. Bishop et al., 2008a). The wavelengthof the band center for this ∼1.9 mm band indicates that anotherhydrated component could be present. Given the trend observedin the mosaics (Figs. 1–2) and the transect images (Figs. 3–5) thatferric oxide/hydroxide signatures are correlated with both theFe/Mg-smectite component and the Al/Si-rich component, ahydrated mineral such as ferrihydrite could be present as wellthroughout the clay-bearing units (and is consistent with themodel composition proposed by Poulet et al., 2008). The H2Ocombination band occurs at 1.93 mm in ferrihydrite (e.g. Bishopand Murad, 2002) and could be contributing towards this band inthe Fe/Mg-smectite region and shifting it towards 1.92 mm. Spec-trum 2 exhibits broader features near 1.4, 2.3 and 2.4 mm that areconsistent with Fe/Mg-smectite and may include a combination ofsaponite or hectorite type structures (both Mg-smectites) mixed

Fig. 4. 3D CRISM images and transect spectra of FRT0000AA7D prepared as in Fig. 3.


with the nontronite type due to the broadness of the features.Spectrum 1 has a strong positive slope from 1.2–2.2 mm in Fig. 6aas well as features consistent with nontronite. In addition there isa shoulder feature near 2.23 mm that could be due to combinationbands (stretching plus bending) of AlFe(OH)2 sites in the smectitestructure. Alternatively, this could be an indicator for acid altera-tion of the nontronite (Madejova et al., 2009).

An additional broad band near 2.5 mm is observed in manyspectra of this unit (e.g. spectra 1 and 3) and could be due to aserpentine (e.g. lizardite) or analcime (a type of zeolite). Thespectra of zeolites such as analcime could be contributing to thesespectra, but cannot be uniquely identified because of the weak andbroad nature of the ∼2.5 mm band and the lack of a band near1.8 mm. Analcime forms in a variety of geochemical environmentsat a pH generally above 11 and under temperatures from ∼100–400 1C (e.g. Boles, 1971; Chipera and Bish, 1999). Analcime canbe formed via sedimentary processes, as well as by alteration ofbasaltic glass, aluminosilicate gel and other zeolites.

Six examples of the Al-phyllosilicate and amorphous Al/Sihydrated type spectra with a band near 2.2 mm due to a combina-tion stretching and bending band for OH connected to Al or Si areshown in Fig. 6b. Spectra 1, 2, 3 and 6 have deeper bands at 1.9 mm(H2O) than at 2.2 mm (OH), which is consistent with Al-smectites,allophane, and opal-CT. Montmorillonite typically has a combina-tion band near 2.20–2.21 mm, while this feature occurs near 2.17–2.19 mm in beidellite spectra and near 2.19–2.20 mm in beidellite-montmorillonite assemblages (Bishop et al., 2011a). Spectrum2 has a broad OH overtone from 1.38–1.41 mm, an H2O combination

band at 1.92 mm and an OH combination band at 2.20 mm that aresimilar to the features characteristic of allophane (Bishop et al.,2013).

Spectrum 3 has a broad OH overtone from 1.38–1.40 mm and anOH combination band at 2.21 mm that is broadened towards longerwavelengths as in the spectrum of opal-A. Spectrum 6 has an OHovertone near 1.41 mm and an OH combination band near 2.21 mmthat is broadened towards longer wavelengths as in opal-CT.Variations in Si-OH/H2O features are due to differences in hydra-tion level and H-bonding for hydrated silica present in alteredglass, opal-A and opal-CT (Anderson and Wickersheim, 1964). Notethat opal-A and opal-CT are similar forms of hydrated silica whereA refers to amorphous, C to cristobalite and T to tridymite (e.g.Guthrie et al., 1995; Dyar and Gunter, 2007). These studies showedthat crystallite sizes are generally 10–30 nm for opal-CT andsmaller for the less mature opal-A.

Spectrum 4 has a deeper band near 2.2 than near 1.9 mm andthe ∼2.2 mm band here is a doublet near 2.17 and 2.20 mm, which isconsistent with kaolin-group minerals (kaolinite, dickite andhalloysite). Subtle changes are observed in the doublet spacingnear 1.4 and 2.2 mm for kaolinite, dickite and halloysite, but thesespecific minerals are difficult to uniquely assign in CRISM data dueto mixing with the other components in the Martian rocks.Kaolinite and dickite are two polytypes of the kaolin groupminerals with different stacking orders (e.g. Dyar and Gunter,2007). Halloysite is a hydrated mineral in the kaolin group and itsspectra include a H2O combination band at 1.91 mm, whereaskaolinite and dickite spectra should not have this feature. The

Fig. 5. 3D CRISM images and transect spectra of FRT0000863E prepared as in Fig. 3.


presence of an absorption feature at 1.92 mm in spectrum 4 eitherindicates the presence of halloysite or of kaolinite or dickite mixedwith a hydrated phase.

Amorphous hydrated minerals such as allophane or ferrihydrite(ferric oxyhydroxide) could be mixed with other minerals to shiftthe H2O combination band to 1.92 mm rather than 1.91 mm that ischaracteristic of smectites, opal and zeolites. A recent study of thespectral properties of mixtures of kaolinite, montmorillonite, opaland altered glass showed that mixtures of kaolinite or montmor-illonite and altered glass best matched the spectral featurestypically observed for the Al/Si-rich unit of those investigated(McKeown et al., 2011), although they were not entirely consistentwith the observed Mawrth Vallis spectra and additional mixtureexperiments are needed.

The drop in reflectance near 2.4–2.5 mm observed in spectra1 and 2 is also found in spectra of zeolites such as heulandite andscolecite (Fig. 6b). The shape of the heulandite spectrum near2.4 mm is similar to that of spectra 2 and 6, while the steeper dropin reflectance near 2.4 mm observed in the spectrum of scoleciteis more similar to that observed in spectrum 1. Isolated regions

characteristic of zeolites have not yet been observed at MawrthVallis, but zeolites may be present in mixtures. Heulandategenerally forms at near neutral pH, rather than the elevated pHsmore typical of other zeolites (e.g. Boles, 1971; Cotton, 2008).Scolecite is less likely to be present as it generally forms in Marineenvironments (e.g. Coombs et al. 2005).

Four examples of spectra with a doublet feature observed inimage FRT0000863E are shown in Fig. 6c. These spectra includebands near 1.39–1.41, 1.92 and 2.17–2.21 mm as in spectra of theAl/Si-rich materials observed in Fig. 6b. However, an additionalband is present near 2.27 mm and a drop in reflectance is observednear 2.4 and 2.5 mm. These spectral features can be explained by afew different scenarios. Spectra of acid-treated Fe-smectites(Madejova et al., 2009) exhibit a doublet near 2.21–2.23 and2.26–2.27 mm (Fig. 6c) and provide a possible explanation for theobserved doublet features in some of the upper Al/Si-type units inMawrth Vallis. This would be consistent with acidic conditionsoccurring following the emplacement of the Fe/Mg-smectite andalteration of this material. The ∼2.27 mm band is attributed to anOH combination (stretch+bend) band and is also consistent with

Fig. 6. Relative CRISM I/F spectra from 1.0–2.65 mm from images FRT0000863E, FRT000089F7, and FRT0000AA7D compared with laboratory reflectance spectra (fromprevious studies, e.g. Bishop et al. 2002a, 2002b, 2005, 2008a, 2008b, 2011a, 2013, Madejova et al., 2009, and the USGS spectral database: http://speclab.cr.usgs.gov/, Clarket al., 2007). (a) ratioed CRISM spectra from four locations exhibiting bands near 2.3 and 2.4 mm (1: x197y239/y270 10�10 from FRT0000863E, 2: x558y264/y76 10�10from FRT0000AA7D, 3: x254y84/y125 10�10 from FRT0000863E, 4: x365y201/y316 5�5 from FRT000089F7) compared with reflectance spectra of Fe- and Mg-richphyllosilicates and analcime, offset for clarity. (b) ratioed CRISM spectra from 5�5 spots in FRT0000863E at six locations exhibiting a band near 2.2 mm (1: x515y176/y312,2: x581y162/y369, 3: x580y285/y371, 4: x203y50/y110, 5: x152y217/y266, 6: x596y237/y253) compared with reflectance spectra of Al-rich phyllosilicates, zeolites andamorphous Al/Si-rich minerals, offset for clarity. (c) ratioed CRISM spectra from image FRT0000863E showing a doublet feature near 2.23 and 2.27 mm (1: x457y138/y30810�10, 2: x515y176/y312 5�5, 3: x581y162/y369 5�5, 4: x447y105/y253 10�10) compared with reflectance spectra of OH-bearing minerals and acid-treated smectite(ATS), offset for clarity.



jarosite (Fe–OH, sulfate) and gibbsite (Al–OH, hydroxide). Otherhydroxide-bearing sulfates such as butlerite exhibit an OH bandnear 2.26–2.27 mm and a drop in reflectance near 2.4 mm. Thus,another possible explanation for the observed doublet feature is amixture of opal, allophane or halloysite with gibbsite (Al[OH]3) oran OH-bearing sulfate. Two of the scenarios to explain this doubletfeature (acid-alteration of smectite or sulfate minerals mixed withthe Al/Si-rich materials) indicate that acidic conditions were likely.

Present at the boundary of the lower Fe/Mg-smectite unit andthe upper Al/Si-rich unit is a region whose spectra contain apositive slope from ∼1.0–1.1 towards 1.8–2.1 mm that is attributedto ferrous iron (Bishop et al. 2008a). Spectra shown in Fig. 7

Fig. 7. CRISM I/F spectra from image FRT0003BFB highlighting change in spectral slopeFRT0003BFB (R D2300, G OLINDEX, B BD2210 parameters) over MOLA topography, 10(b) CRISM I/F spectra in one column including several spectra: 1- with 2.20 mm (dark bland a weak positive slope from 1–1.8 mm characteristic of Fe2+ in several minerals (lightdoublet near 2.23 and 2.27 mm and a strong positive slope from 1–1.8 mm spectra (olivphyllosilicates and a strong positive slope from 1–1.8 mm spectra (yellow), 6- with band1–1.8 mm (orange), 7- with bands at 2.29 and 2.40 mm due to Fe/Mg-phyllosilicates, andratios. (c) ratioed CRISM I/F spectra corresponding to those shown in (b). Note that theattributed to dust or sand coatings at the site of the denominator spectrum. (d) labo0.3–3.3 mm: olivine, celadenite (Cld), chamosite (Chm), glauconite (Glt) and siderite. (Forto the web version of this article.)

illustrate this change in slope across the boundary of the lowerFe/Mg-smectite and upper Al/Si-rich units. I/F spectra are shownin Fig. 7b for spectra collected along a transect in a single columnfrom the image shown in Fig. 7a. Ratio spectra were prepared fromthe I/F spectra in Fig. 7b by dividing each of them by a spectrallyneutral region (spectrum 8). The resulting ratio spectra aredisplayed in Fig. 7c and are offset to illustrate their verticalposition in the stratigraphy. Nontronite-type spectra (6, 7) and anontronite plus ferrous phase spectrum (5) both have a band near2.29 mm attributed to the OH combination band in nontronite.Similarly, a spectrum consistent with Al-smectite or an amorphousAl/Si hydrated material (spectra 1 and 2) contains a feature near

from 1.0–1.8 mm attributed to a ferrous component. (a) 3D surface of CRISM imageX vertically exaggerated, white line marks transect where spectra were collected.ue) due to Al-OH in phyllosilicates and/or Si-OH in hydrated silica, 2- with 2.20 mmblue), 3- with 2.20 mm and a strong positive slope from 1–1.8 mm (cyan), 4- a weake), 5- with bands at 2.29 and 2.40 mm due to Fe-OH (and maybe some Mg-OH) ins at 2.29 and 2.40 mm due to Fe/Mg-phyllosilicates and a weak positive slope fromfinally the spectrum of a spectrally neutral region used as the denominator for theratioed spectra often have some degree of positive slope due to a negative sloperatory reflectance spectra of several Fe2+-bearing minerals for comparison frominterpretation of the references to color in this figure legend, the reader is referred


1.9 and 2.20 mm and a similar spectrum (3) contains these featuresplus an enhanced positive slope from 1–1.8 mm attributed to aferrous component. This enhanced positive slope is caused by abroad band arising from an Fe2+ electronic transition in phyllosi-licates (e.g. Burns, 1993; Lear and Stucki, 1987) and is mapped ingreen for the L images (Figures 1-5). Fe2+–Fe3+ charge transfertransitions are also observed near 0.7-0.76 µm for phyllosilicates(e.g. Faye, 1968; Anderson and Stucki, 1979; Lear and Stucki, 1987).Spectra 3 and 4 that exhibit this enhanced slope from 1–1.8 mmalso have weak features near 2.29 mm indicating there may be amixture of the Al/Si-rich species and the Fe/Mg-rich species.

Shown in Fig. 7d are several spectra of Fe2+-bearing minerals.Phyllosilicates such as glauconite (Fe2+Mg–mica), chamosite (Fe2+

-rich chlorite with some Mg) and celadonite (Mg-rich chlorite withsome Fe2+) often form in association with smectites and are thusarguably more likely to be present in between two phyllosilicateunits than would be unrelated minerals such as olivine or Fe–carbonate. Other possibilities are that some of the Fe3+ in thenontronite structure was reduced to Fe2+or that some Fe2+ wasretained in the nontronite structure at the time of formation.Reduction of nontronites has been performed in the lab usinginorganic and biogenic processes (e.g. Lear and Stucki, 1987;Stucki, 2006), as well as by heating (Morris et al., 2009). Reductionof Fe3+ in nontronite also causes deprotonation of the OH bound tooctahedral cations and decreases the intensity of the OH spectralfeatures (Manceau et al., 2000; Fialips et al., 2002). Changes in theOH vibrational energies are also expected when Fe3+ is reduced toFe2+ or when a combination of both Fe3+ and Fe2+ are present (e.g.Bishop et al., 2002a; Neumann et al., 2011). Manceau et al. (2000)observeda change in the Fe2OHstretching vibration from3570 cm−1

for Fe3+ to 3623 cm−1 for Fe2+. The Fe2OH bending vibrations near844 and 822 cm−1were found to shift to 814 cm−1 upon reduction ofthe iron for up to 1 h and then finally to 653 cm−1 upon 4 h ofreduction (Fialips et al., 2002). They noted that this is similar to theFe2+2 OH bending vibrations in talc (Wilkins and Ito, 1976) andsuggest that trioctahedral domains of Fe2+ were formed in theoctahedral layer upon reduction of nontronite. Adding the stretch-ing and bending vibrational energies (3623 and 653 cm−1 fromliterature cited above) for Fe2+2 OH as in the Bishop et al. (2002a)study gives 4276 cm−1 or 2.339 mm for the OH combination band forFe2+ clusters in reduced nontronite. This value is similar to the OHcombination bands observed in spectra of Fe2+-bearing chlorites (e.g. chamosite) andmicas (e.g. biotite, glauconite). One inconsistencywith the presence of an Fe2+ clay is that the Fe2+OH combinationbands near 2.33–2.34 mm are not observed. However, reflectancespectra of laboratory mixtures of nontronite and chlorite showedthat the nontronite features dominated the spectral signature near2.3 mm for up to 50 wt% chlorite (Saper and Bishop, 2011). Thus, anFe2+ bearing clay could be present as a prominent but not majorcomponent of this unit characterized by an increasing slope from 1–2 mm. A ferrous phase could also have been formed due to a redoxprocesswhen the upper Al/Si-rich unit formed that then intermixedwith each of the phyllosilicate units near their boundary.

A scene from CRISM image FRT000089F7 over HiRISE is shownin Fig. 8 to illustrate the transition from Fe/Mg-smectite to theupper Al/Si-rich unit. Differences in texture for these units havebeen observed (McKeown, 2010); the Fe/Mg-smectite unit exhibitsfracturing and polygonal features on a larger scale (∼1–5 m) thanthe Al/Si-rich unit (∼0.5–2 m).

Shown in Fig. 9 is a summary of the clay profile at MawrthVallis. This shows Fe3+/Mg-smectite at the bottom in the oldestunit. Spectra of this unit are similar to nontronite or Mg-richnontronite. A ferrous phase is present at the transition betweenthe Fe3+/Mg-smectite and the upper Al/Si-rich unit (Al-smectite,kaolin-group minerals, hydrated amorphous Al/Si minerals andpossibly gibbsite, Fe–OH-bearing sulfates and/or acid-treated

clays). This ferrous phase cannot be uniquely constrained, but islikely due to reduction of some of the Fe3+ in the nontronite orformation of another Fe2+ phase at the boundary of the majorphyllosilicate units. The upper Al/Si-rich unit contains hydratedamorphous species (e.g. opal, altered silica glass, allophane)typically in thick deposits and often mixed with kaolinite and/ormontmorillonite from above and sometimes mixed with the ferrouslayer below. Kaolinite and montmorillonite are present in the upper(youngest) units, usually in small outcrops for the most pureexposures. This upper Al/Si-rich unit is likely a mixture of all ofthese components (McKeown et al., 2011) and may have somesulfates containing OH and Fe such as jarosite or butlerite or acidalteration products. Ferric oxide-bearing materials are associatedwith both the lower Fe3+/Mg-smectite unit and the upper Al/Si-richunit. The clay-rich unit is covered by a caprock, which is a fewmeters-thick where not eroded away.

4. Discussion of data

The phyllosilicate outcrops observed in the Mawrth Vallisregion that are described here and in previous studies (Pouletet al., 2005; Loizeau et al., 2007; Michalski and Noe Dobrea, 2007;Bishop et al., 2008b; Mustard et al., 2008; McKeown et al., 2009;Loizeau et al., 2010; Noe Dobrea et al., 2010) indicate thatabundant water was present here during the Noachian period. Inthis study we discuss factors that could have shaped the formationand alteration of the observed phyllosilicates as well as implica-tions of these phyllosilicate-bearing rocks for aqueous processes,chemistry, biology, and preservation of potential signatures ofthese. Phyllosilicates have been detected in a number of outcropsof Mars' ancient crust surrounding Mawrth Vallis (Noe Dobreaet al., 2010). Phyllosilicates have also been identified across theplanet in ancient rocks at numerous sites (e.g. Poulet et al., 2005;Mustard et al., 2008; Murchie et al., 2009b; Wray et al., 2009;Carter et al., 2010) suggesting that the processes forming phyllo-silicates here may have been widespread and that only a fractionof the phyllosilicate-rich rocks have been exposed on the surface.

4.1. Possible formation processes for clay minerals at Mawrth Vallis

Analysis of CRISM and OMEGA spectra and HRSC and HiRISEimages across the Mawrth Vallis region reveal a thick (∼100–200 m deep) deposit of Fe/Mg-smectite that has been covered by athinner (∼50 m deep) unit of Al-phyllosilicate and hydrated silica(Poulet et al., 2005; Loizeau et al., 2007; Bishop et al., 2008b;Mustard et al., 2008; Wray et al., 2008; McKeown et al., 2009; NoeDobrea et al., 2010; Michalski et al., 2010). A common stratigraphy(Fig. 9) has been observed across the Mawrth Vallis region wherean Fe2+ phase is often present at the boundary of these materials(Bishop et al., 2008b; McKeown et al., 2009). Spectra shown inFig. 7 are typical of this region and include examples of outcropscontaining Fe/Mg-smectites, Al-smectites, amorphous hydratedAl/Si phases, kaolin family minerals, and perhaps acid alterationproducts. The hydrated silica and allophane-like componentsappear to be mixed with the Al-phyllosilicates in many localities.

Alteration of basaltic rocks on Earth typically produces Al- andFe-bearing dioctahedral smectites such as montmorillonite andnontronite in high water/rock ratio environments with moderatesilica activity (Velde, 1995; Caillaud et al., 2006). The abundantFe/Mg-smectite found on Mars in regions like Mawrth Vallislikely formed via aqueous alteration of mafic to ultramafic rocks(Chevrier et al., 2007; Meunier et al., 2010). Fe/Mg-smectites arecommonly formed in marine, lacustrine, and hydrothermal sub-marine environments on Earth and the different mineralogiesformed in each of these environments can provide information

Fig. 8. (a) portion of 3D CRISM FRT000089F7 over HiRISE, 3X vertical exageration (R dD2300, G OLINDEX, B BD2210) with yellow box indicating location of image in (b)),(b) partial view of HiRISE image ESP_021510_2040_COLOR illustrating surface morphologies of the phyllosilicate-bearing units with blue and red boxes indicating locationsof images in (c) and (d), (c) HiRISE surface view of Al/Si-rich unit, (d) HiRISE surface view of Fe/Mg-smectite unit. (For interpretation of the references to color in this figurelegend, the reader is referred to the web version of this article.)


about the geochemical formation environment (e.g. Chamley,1989; Velde, 1995). Fe-rich saponite (Fe/Mg-smectite) commonlyforms in sediments derived from basaltic volcanic material(Desprairies et al., 1989; Parthasarathy et al., 2003), while morereduced smectites such as Fe2+-stevensite are more typical inhydrothermal brines (e.g. Badaut et al., 1985). Under low-temperature (e.g. o20 1C) and low-oxygen (dysoxic) conditionsFe-rich smectites form under neutral to slightly basic conditionsusing Fe2+ in solution (Harder, 1976). Harder (1976) found that Fe2+

was necessary for low-temperature formation of the brucite-gibbsitetype octahedral layers of the clay minerals and that the Fe2+ couldthen be oxidized to Fe3+ as the mineral formed. More recently,synthetic nontronites were formed under hydrothermal conditions(150 1C, pH∼12) from both ferric chloride and ferrous chloride(Andrieux and Petit, 2010). Farmer et al. (1994) synthesized non-tronite at 90 1C and found that reduced conditions during the earlyphase of synthesis produced better crystallized products. Decarreauet al. (2008) also used a ferrous starting gel to form synthetic

Fig. 9. Example stratigraphy of phyllosilicate-bearing units. A unit spectrallydominated by nontronite and Fe/Mg-smectites and also containing ferric oxides/hydroxides/oxyhydroxides (FeOx) was emplaced first and is at least 100–200 mthick in the Mawrth Vallis region. This is covered by a thin (5–10 m) unit containingFe2+ phases that is most frequently mixed with the lower Fe/Mg-smectite unit, butis sometimes mixed with the upper Al/Si-bearing unit (10–50 m). This upper unitcontains a mixture of minerals and other materials including montmorillonite,kaolinite/halloysite, hydrated silica, allophane, and acid-treated materials as wellas an FeOx component. Finally, caprock is present in some regions along thesurface.


nontronites and produced crystalline nontronites at temperaturesover the range 75–150 1C. Cuadros et al. (2011) analyzed natural Fe-rich smectites formed under low-temperature submarine hydro-thermal conditions via slow reaction of Fe3+ hydroxides, detritalsilicates and silica. This study indicates that poorly crystalline Fe-richsmectites containing Al and/or Mn as well as Fe in octahedral sitescould form via this mechanism.

Abundant Fe3+-bearing nontronite forms in dysoxic and anoxicregions in the ocean on Earth where oxygen is depleted dueto reaction with Fe2+ released into solution from basaltic rocks(Bischoff, 1972; Arrhenius, 1986). Further, smectites are well-known as the dominant clay mineral forming in deep-sea envir-onments where volcanic activity is prevalent (e.g. Odin, 1986).Once the nontronite forms, some oxidation is necessary to produceFe3+ and stabilize the nontronite (Lonsdale et al., 1980). Odin(1986) also reports formation of authigenic glauconitic smectitecontaining both Fe2+ and Fe3+. The oxygen-poor atmosphere onthe early Earth, similar to past and present Mars, supportedformation of Fe-rich smectites such as nontronite from basaltic-composition ash in sedimentary environments (Harder, 1988).Thermodynamic considerations support formation of smectiteson Mars only if liquid water was present; solid or vapor phase H2Ocould not have supported formation of smectites (Gooding, 1978).The silica content in solution is also important for Fe-smectiteformation, such that lower Si/Fe ratios give hydroxides and onlypoorly crystalline nontronite, while higher Si/Fe ratios produceonly amorphous products (Harder, 1976).

This Fe/Mg-smectite unit exhibits layered textures consistentwith ash as the basaltic precursor (e.g. Bishop et al., 2008b).Sedimentary processes may also have taken place (Michalski andNoe Dobrea, 2007; Bishop et al., 2011b). Alteration could haveoccurred throughdirect deposition of ash into abodyof liquidwater,or via subsurface ground water (McKeown et al., 2009; Ehlmannet al., 2011). Alternative sources include water-lain basaltic sedi-ments and can involve impact ejecta as a minor contribution. Thelayering is complex indicating perhapsmultiple depositional eventsover a prolonged period of time (Michalski and Noe Dobrea, 2007).Other possible formation processes such as pedogenesis and hydro-thermal alteration are discussed by McKeown et al. (2009) and NoeDobrea et al. (2010) and could have led to the observed claystratigraphy. Subsequent aqueous alteration and leaching of Fe andMg from the smectite or ash precursormaterial could have producedamorphous hydrated Al/Si-rich phases, Al-smectite, Al/Mg-richmica,and kaolin-group minerals. Alternatively, a change in the aqueouschemistry (e.g., hydrothermal activity ormore acidic conditions) or achange to a less basaltic (more Si-rich) volcanic ash or sedimentprecursormaterial couldhaveproducedtheAl/Si-richupperclayunit.

Another possible scenario is alteration of interbedded source rockswith different chemistries, resulting in changing alteration profilesfromFe/Mg-smectites inmoremafic toultramafic lowerrockstomoreSi-rich upper rocks.

What happened in the intermediate strata is less certain andperhaps more interesting geochemically, especially with regard topotential habitability conditions. One model under considerationhere is that the nontronite unit formed a hard, resistant and non-porous surface. If this occurred then the ash/sediments depositedon top of the nontronite could have been altered in one or morelong-term and wide-spread aqueous events that produced exten-sive leaching and alteration near the surface and formed kaolinite/halloysite and other Al- and Si-rich clays. The Fe2+ and K leachedout of the ash/sediments could have been trapped at the imperme-able nontronite border and redeposited to form ferrous mica.Another possibility is that under long-term exposure to aqueousconditions smectites can convert to Fe2+-phyllosilicates such asglauconite or stevensite if microbial activity or organic compoundsprovide a reducing environment, if wet/dry cycling occurs, or inthe presence of abundant iron or high salinity (Chamley, 1989;Nagy, 1995). These possible formation processes for the clay-richunits at Mawrth Vallis are summarized in Fig. 10.

Velde (2003) describes organic matter as the most widelyavailable reducing agent for smectites. Stucki (1997) reports NH4

fixation coupled with Fe reduction in smectites. Huggett andCuadros (2005) observed illitization of smectite in silty clay-bearing soils at the Headon Hill Formation (Isle of Wight) underwet/dry cycling environments; however, the Fe-reduction wasbacterially mediated. A related site at this formation contains soilsrich in smectite, illite, kaolinite and chlorite, covered by a layer ofCa sulfate attributed to an hypersaline environment, and toppedwith an Fe2+-rich illite-smectite glauconite unit (Huggett andCuadros, 2010). This study identified two mechanisms for Fe3+

reduction and illitization of the soils: hypersaline clay authigenesisand wet/dry cycling with microbial activity. As Ca sulfates havebeen identified at Mawrth Vallis (Wray et al., 2010), the Fe3+

reduction mechanism under hypersaline conditions could be aviable mechanism for the Fe2+ material here. Fe3+ reduction viawet/dry cycling is a less likely mechanism for Mars as this appearsto require microbial activity.

Lab experiments suggest that mixtures of Fe2+-phyllosilicatesand nontronite would exhibit a spectral band near 2.3 mm consis-tent with a nontronite mixture containing ≤25 wt% Fe2+-phyllosi-licates, but also an increased spectral slope characteristic of ferrousmaterials (Saper and Bishop, 2011). Thus, we might not observespectral evidence at 2.3 mm for the Fe

2+-phyllosilicates even at

abundances large enough to induce the ferrous spectral slope.Neutral to alkaline conditions favor smectites under surface

temperatures and pressures on Earth in the presence of Ca andfavor mica in the presence of K, whereas acidic conditions supportformation of kaolinite and hydrated silica (Jackson, 1959; Chamley,1989). More strongly alkaline conditions support the formation ofzeolites (Kawano and Tomita, 1997; Hall, 1998). Al-smectites andopal are commonly found in altered basaltic environments, includ-ing acid-leaching environments. A drop in pH over time due to areduction in liquid water and concentration of salts on the surfacecould have supported formation of aluminosilicate products andiron hydroxides rather than Fe-rich smectites (e.g. Harder, 1976).Geochemical modeling by Chevrier et al. (2007) also suggestsFe2+-bearing chlorites would form instead of nontronite undersaline conditions. Thus, leaching might not be necessary to formthe observed Fe2+-rich unit at the boundary of the Fe/Mg-smectiteand the upper Al/Si-rich minerals. If the latter aqueous eventforming the Al/Si-rich unit occurred under lower pH conditionsand followed the Chevrier model, then Fe2+ chlorite could havecrystallized out first with Al/Si-rich species forming later.

Fig. 11. Martian phyllosilicate timeline. This diagram illustrates how phyllosilicatesmay have been formed in aqueous environments on the early Mars (prior to 4 billionyears ago) and that this may have been a pervasive processes. Subsequently, much ofthe phyllosilicate-bearing unit was covered and the water disappeared over time.Phyllosilicates are currently only visible where the caprock has been eroded away,although they may be wide-spread below the surface.

Fig. 10. Formation diagram for phyllosilicate units illustrating several likely steps that occurred: (1) basaltic ash or sediments were present in a liquid water reservoir duringthe Noachian period, (2) a thick unit containing substantial Fe/Mg-smectite was formed through alteration of the ash/sediment, (3) a subsequent aqueous event occurredthat formed a mixture of Al/Si-rich alteration products and may have involved leaching or redox reactions at the boundary with the Fe/Mg-smectite-rich unit, (4) caprocklikely from lava and basaltic sand covered all the clay-bearing units.


Meunier (2007) describes formation of hydroxyl-interlayerminerals in smectites that appear as polymers of Fe(OH)3 or Al(OH)3 clusters. Given the presence of reducing conditions oradsorbed organics in the interlayer regions, these Fe3+(OH)3polymers could be reduced to form Fe2+(OH)2 clusters. Meunieret al. (2010) postulate that M2+(OH)2 clusters with M¼Fe2+, Mg2+

could have formed similar polymer chains in smectites under earlyEarth conditions. Thus, this same mechanism is plausible for earlyMars and provides another explanation for formation of an Fe2+

phase associated with some Fe/Mg smectite units on Mars, althoughFe2+(OH)2 species would likely at least partially re-oxidize on thesurface.

4.2. Availability of water during and after formation of thephyllosilicate-bearing material

Phyllosilicates generally form under high water/rock ratio environ-ments, and high liquid water availability as a solvent during depositioncontrols the type of phyllosilicates formed (e.g. Tosca and Knoll, 2009).Smectites are common in regions dominated by wet/dry cycling of theclimate (e.g. Chamley, 1989). The diagram in Fig. 10 depicts a scenariowhere clays may have formed in an aqueous environment on earlyMars. These clay-rich units would remain after the liquid water is nolonger present and would likely be buried over time by othermaterials. Fe/Mg-smectite is not only observed at Mawrth Vallis, butacross the planet in multiple outcrops (e.g. Murchie et al., 2009b;Carter et al., 2011). Formation of a planet-wide Fe/Mg-smectite unitmay have occurred early in the planets' history that was then eroded,buried or altered depending on local geologic activity. This possiblywide-spread Fe/Mg-smectite unit is observed currently in smallerexposures on the surface as illustrated in Fig. 11, a schematic of apossible global geologic history.

Water availability after deposition will control the types ofminerals present as well and hydrated silica is an importantindicator of water availability (Tosca and Knoll, 2009). When waterinteracts with mafic rocks, the mobile silica is easily dissolved tobe later redeposited (McLennan, 2003) and hydrated silica hasbeen observed in many studies of altered volcanic material (e.g.Morris et al., 2003; Bishop et al., 2005). Hydrated/amorphous silicaconverts to opal-CT and then to quartz in water (Siever, 1962).Tosca and Knoll (2009) estimate this process takes on the order of

100 million years. Hydrated silica has been detected on Mars inassociation with both clay minerals and sulfates (e.g. Bishop et al.,2008b, 2009; Milliken et al., 2008; Murchie et al., 2009a, 2009b).Many of the silica deposits associated with sulfates have beendetected in Hesperian aged rocks (e.g. Milliken et al., 2008).Analyses of the global TES dataset indicate detections of quartzor Si-rich components in a few small areas (Bandfield et al., 2004).More recent analyses of TES data suggest that amorphous Al/Sicomponents could be present in many regions of Mars (Ruff, 2004;Michalski et al., 2006; Rampe et al., 2011). Regardless of whetherquartz is present in some locations on Mars, because amorphoussilica still persists in many regions, water availability after deposi-tion of this silica must have been limited.

The availability of subsurface water on Mars can also be assessedthrough smectite diagenesis. Conversion of smectite to illite orchlorite is dependent on temperature, burial history and theavailability of water. Tosca et al. (2008) attempted to constrain thisin a general way by modeling the effects of time and temperatureon sediment diagenesis. They concluded that smectites deposited∼3.5 Ga and buried to at least 300–400 m depth should haveconverted to chlorite and/or illite in the presence of water and K+.Because thick Fe/Mg-smectite units have persisted at Mawrth Vallisand elsewhere on Mars, liquid water has likely been limited since


their deposition or they have remained relatively close to thesurface.

Briefly examining the geomorphic evidence for liquid water,several types of channels are observed on Mars' surface. Dendriticchannels, which generally form due to surface runoff, occur primarilyin Noachian-aged terrains (e.g. Pieri, 1976; Squyres, 1989; Harrisonand Grimm, 2005). The rate of valley formation declined through theHesperian, transitioning to outflow channel formation, and wasdominated by groundwater sapping rather than surface runoff(Harrison and Grimm, 2005 and references therein). Channel forma-tion continued to slow and eventually stopped in the early Amazo-nian, with few examples of Amazonian-aged channels (Harrison andGrimm, 2005). Furthermore, the rate of crater degradation appears tohave decreased from the late Noachian to the early Amazonian,indicating there were more modifying forces, such as surface orsubsurface water, at work earlier in Mars' history (Schultz, 1986;Pollack et al., 1987). The mineralogic and geomorphic evidence bothsupport the presence of more abundant liquid water on Mars in theNoachian and early Hesperian periods.

4.3. Implications of phyllosilicate-bearing rocks as reactants for pre-biotic chemistry and/or habitats for microbes

In this section we discuss the structure of smectite clays and thetypes of chemical reactions that have been performed on smectitesurfaces. Numerous studies have investigated organic chemistry onmontmorillonite and other smectites, as detailed below. We alsosummarize interactions of microbes with clay minerals.

4.3.1. Smectite structureThe structures of phyllosilicate minerals are well-known (e.g.

Suquet et al., 1975; Bailey, 1980; Güven, 1988; Deer et al., 1992)and are classified based on: (1) the ratio of tetrahedral tooctahedral sheets, (2) dioctahedral or trioctahedral occupancy ofthe octahedral sheet, (3) the charge expressed at the interlayer siteresulting from cation occupancies in the octahedral and tetrahe-dral sheets, and (4) the occupancy of the interlayer region.Smectites are composed of octahedral (O) sheets bonded totetrahedral (T) sheets and have a 2:1 ratio with an interlayerregion (I) in between each T–O–T unit as summarized recently byBishop et al. (2008a) and shown in Fig. 12. The general smectiteformula follows IM2−3□1−0T4O10(OH)2, where:

�
I is the interlayer, which is occupied by a cation (e.g. K+, Na+,Ca2+, NH4
+, H3O+);
� M is an octahedral cation (typically Al3+, Fe3+, Mg2+, or Fe2+,
but occasionally also others such as Li, Mn, Zn, Cr, or Ti);
� □ is a vacancy in an octahedral site; �
Fig. 12. Smectite structural model. (a) view of the smectite structure along the aunit cell direction with Si tetrahedra shown in yellow, M octahedral shown in green

T is the tetrahedral cation, commonly Si and Al, and sometimesFe3+;

(M¼Al, Fe, Mg), O atoms in red, OH groups in orange and H O molecules in blue.
� 2
The structure also includes interlayer cations (not shown) and the sheets expandupon hydration to incorporate additional water molecules in the interlayer region.(b) a similar view of the smectite structure illustrating an expanded interlayerregion where organic molecules can react with clay mineral surface. A carboxylicacid, phenol and alanine are shown as examples. (For interpretation of thereferences to color in this figure legend, the reader is referred to the web versionof this article.)

the subscripts 2–3 and 1–0 refer to variations in the cationoccupancy of the octahedral sites that depend on whetherthe smectite is dioctahedral (two M3+ cations) or trioctahedral(three M2+ cations).

The 2:1 smectite group includes the Al-rich dioctahedralminerals montmorillonite and beidellite, the Fe-rich dioctahedralminerals nontronite and ferruginous smectite, and the Mg-richtrioctahedral minerals saponite, hectorite, and stevensite. Syn-thetic Fe/Mg-smectites (Grauby et al., 1994) and natural Fe/Mg-smectites (Post and Plummer, 1972; Bishop et al., 2008b) havebeen found and likely exist with pockets of dioctahedral andtrioctahedral character. Cation substitutions in the smectite group(e.g. Al3+ for Si4+ or Mg2+ for Al3+) cause a slight negative chargethat is balanced by small numbers of cations, such as Mg2+, Ca2+

and Na+, in the interlayer site along with varying amounts ofassociated H2O. Smectites contain water bound to cations in theinterlayer region that includes structural and outer sphere waterand is distinct from the adsorbed water present on most mineralsurfaces (e.g. Fripiat et al., 1960; Russell and Farmer, 1964; Farmerand Russell, 1971; Bishop et al., 1994). This interlayer water iscomplex in smectites because the inner sphere H2O molecules


(directly H-bonded to the cation) and outer sphere H2O molecules(not directly H-bonded to the cation, but to other H2O molecules)exhibit different vibrational energies due to differences in H-bond-ing, proximity to the interlayer cation, and tetrahedral surfacecharge (e.g. Fripiat et al., 1960; Russell and Farmer, 1964; Farmerand Russell, 1971; Bishop et al., 1994).

4.3.2. Chemical reactions on smectite surfacesPhyllosilicate-bearing rocks may have been an ideal place on

Mars for pre-biotic chemistry and possibly the development of lifeas well. The expansive clay deposit at Mawrth Vallis implies a largeand persistent aqueous system (Bishop et al., 2008b). The chan-ging redox conditions and the mildly acidic environment asso-ciated with the hydrated silica and kaolinite/halloysite outcropssuggest an active early chemistry on Mars. The large and variedphyllosilicate system at Nili Fossae (Ehlmann et al., 2008a, 2008b,2009; Mustard et al., 2008) indicates active aqueous processesnear Isidis Basin as well.

Delivery of organic molecules to early Earth by comets andasteroids has been shown by many studies (e.g. Anders, 1989;Chyba et al., 1990; Delsemme, 1997; Pierazzo and Chyba, 1999).Lazcano-Araujo and Oro (1981) estimate that 1023–1025 g oforganic carbon was brought to Earth through meteorite andcomet activity by 3.5 Ga. Kanavarioti and Mancinelli (1990)assume a similar rate for Mars. If only 1% of this organic materialwas not destroyed by impact or geologic activity by 3.5 Ga, thenthere would be substantial amounts of organic material buriedbeneath the Martian surface. Assuming 120 g of organic carbonwas brought to Mars, a Martian surface area of 144,400,000 km2,and a 1% survival of organic material on the surface would give∼700 metric tons of organic carbon per square kilometer. (This ishighly speculative and likely an upper maximum.) Some of thisorganic carbon could have been amino acids and the aminoacid alanine, for example, would have been very stable on Marsat 0 1C based on degradation rate calculations (Kanavarioti andMancinelli, 1990).

Phyllosilicates, especially smectites, can catalyze a variety ofmundane and complex chemical reactions due to the acidity oftheir internal surfaces and by bringing together molecules on theirsurfaces (Pinnavaia, 1983). This results in expanded interlayerregions (Fig. 12b) for “pillared” clays that bear a monolayer ofhydrated polyoxo cations and pillaring cations surrounded bynumerous H2O molecules (e.g. Pinnavaia et al., 1984). Theng(1974) described the adsorptive and catalytic interactions oforganic molecules on clay surfaces early on. More recent experi-ments have shown adsorption and reaction of a variety of organicson clay surfaces (e.g. Dashman and Strotsky, 1982; Hedges andHare, 1987; Kikuchi and Matsuda, 1988; Bergaya et al., 2006). Theparticular features of clay chemistry governing these reactionsinclude the local Bronsted acidity of clay surfaces, shape specificityof reaction sites on clay surfaces, motion restrictions on watermolecules and the binding properties of specific cations (Pinnavaiaand Mortland, 1986). They further suggest that transition metalcations with unfilled d orbitals residing in exchangeable cation sites(e.g. Fe, Mg) create acidic conditions through solvating watermolecules and freeing up H+ in the interlayer region. In order formineral surfaces to act as efficient templates for chemical reactionswith amino acids they must have the ability to adsorb organicmolecules on their surfaces, concentrate these adsorbed species,and activate polymerization of these adsorbants (Lambert, 2008;Meunier et al., 2010). Organic reactions on montmorillonite sur-faces showed that montmorillonite can catalyze formation of RNAand other precursor molecules for the origin of life (e.g. Ferris andHagan Jr., 1986; Ferris et al., 1989; Franchi et al., 2003; Ferris, 2005;Ferris, 2006). Formose-like reactions (forming sugars from

formaldehyde) were performed in different media and silicateminerals were found to stabilize ribose, the sugar in RNA(Vázquez-Mayagoitia et al., 2011). This supports a role for clayminerals in early organic chemistry reactions that could have beenintegral to the origin of life as described by Lawless (1986).Smectites have been found to bind organics much longer thanother clay minerals such as kaolinite (Wattel-Koekkoek et al.,2003). This is likely due to the unique expandable and chargedcharacter of the interlayer region of smectites.

A variety of experiments have been performed on montmor-illonites treated to contain Fe in the interlayer sites (Banin et al.,1997). These Fe-exchanged montmorillonites were found to simu-late well the results of the Viking biology experiments (Banin andRishpon, 1979; Banin and Margulies, 1983). Metal ions in a claymatrix may also have played a crucial role in the origin and earlyevolution of life by altering the behavior of biomonomers such asamino acids and nucleotides (Lawless, 1986). These biomonomerswere adsorbed preferentially onto clay surfaces with Fe or Mg inexchangeable cation sites (Odom et al., 1979; Lawless et al., 1985).More recent studies have also observed increased adsorptionof organic molecules in interlayer sites for smectites containingmetal cations in the exchangeable sites (Franchi et al., 2003).Fe-bearing phyllosilicates such as nontronite, Fe-rich montmor-illonite and glauconite/illite form preferentially in largely anoxicwaters (Harder, 1988) such as the conditions on early Mars. Longpolymers of amino acids and related compounds were synthesizedon montmorillonite and other mineral surfaces (e.g. Ferris et al.,1989; Orgel, 1998), but it remains to be shown if polynucleotidesynthesis catalized by montmorillonite can proceed to create RNAsufficiently to form life (Orgel, 2004).

4.3.3. Clay interactions with microbesIn order to sustain life an environment must provide essential

nutrients, biologically accessible energy and liquid water (Nealson,1997). Studies of terrestrial marine basalts with chemical compo-sitions consistent with Mars (McSween et al., 2009) have shownthat these rocks contain sufficient requirements (e.g. nutrients,water, radiation protection) to sustain life (Fisk et al., 1998). Otherstudies have shown that microbes may even be facilitatingpalagonitization of basalt into clays and other altered phases(e.g. Alt and Mata, 2000; Konhauser et al., 2002; Benzerara et al.,2005). Microbial weathering of Fe-rich phyllosilicates has alsobeen observed (Sanz-Montero et al., 2009).

Experiments with soil bacteria and viruses have shown thatthey can survive in a variety of soil and clay environmentsincluding temperature and moisture extremes replicating Martianconditions (Hawrylewicz et al., 1962; Foster et al., 1978; Moll andVestal, 1992). Elevated growth rates of microbes were noted for aneutral montmorillonite system relative to an acidic one (pH 3.6–4.0) with synthetic Fe3+-exchanged montmorillonite (Moll andVestal, 1992). Organic molecules and metal hydroxides presentin smectite interlayer regions (Fig. 12b) and adsorbed on grainsurfaces may have provided nutrients for these microbes.

4.3.4. Reduction of phyllosilicates: chemical and biologicalStudies of the oxic/anoxic transition in marine sediments have

noted a color transition from tan to green that is associated withreduction of Fe3+ in smectites and other clays (e.g. Lyle, 1983;Drodt et al., 1997, 1998; König et al., 1997, 1999; Badaut et al.,1985). This redox transition could be chemically and/or biologi-cally mediated. Laboratory experiments have shown that Fe3+ inphyllosilicate structures can be reduced via chemical or biologicalprocesses (e.g. Stucki, 2006; Dong et al., 2009). The oxidation stateof the iron in smectites affects a number of properties includinglayer charge, cation exchange and fixation capacity, swelling in


water, particle size, specific surface area, layer stacking order,magnetic exchange interactions, octahedral site occupancy, surfaceacidity, and reduction potential (Stucki et al., 2002). This studyfurther showed that reduction of the octahedral Fe3+ to Fe2+ ismuch more complex than simply transferring an electron in thesmectite structure. The ancillary reactions that occur due to theredox change also produce a large degree of modification tothe phyllosilicate structure (Stucki et al., 2002).

Microbial reduction of Fe3+ to Fe2+ in clay minerals has beenreported in more than 20 studies, summarized by Dong et al.(2009). The microbes participating in these reactions are anaerobicand are commonly found in soils, sediments, and hydrothermalenvironments on Earth. Smectites with expandable layer struc-tures, such as montmorillonite and nontronite, have been foundto provide the highest reduction rates, while clays with non-expanding structures, such as illites and chlorites, afford theslowest reduction rates (Stucki et al., 1987; Gates et al., 1998;Dong et al., 2003, 2009; Stucki, 2006; Jaisi et al., 2007). Elevatedbioreduction levels are also observed for clay minerals with highFe3+ contents (Ernstsen et al., 1998; Gates et al., 1998), highersurface area (Ernstsen et al., 1998; Jaisi et al., 2007), and lowerlayer charge (Dong et al. 2009; Seabaugh et al., 2006).

Studies by Lee et al. (2006) and Stucki (2006) report thatbioreduction of smectites induced small changes in the mineralstructures that were reversible upon re-oxidation. However, otherstudies have observed dissolution of clay minerals upon microbialreduction of Fe3+ resulting in formation of biogenic silica (e.g.Dong et al., 2003; Furukawa and O'Reilly, 2007; Vorhies andGaines, 2009). If Fe is abundant and reduction is complete, thenthe entire smectite structure is destabilized and dissolves (J.Cuadros, pers. comm., 2012). Microbial reduction of Fe3+ insmectites also facilitates illitization (Kim et al., 2004; Zhanget al., 2007), although a number of factors, such as temperature,pH and water/rock ratio control and/or facilitate this reaction aswell (Hower et al. 1976; Whitney, 1990; Drief et al., 2002). Morerecently, Neumann et al. (2011) suggest that smectite illitizationcan be reversible.

Dithionite is among the most common chemical reductantsused for lab experiments of ferric iron reduction in clays (Stucki,2006). Other inorganic reductants include hydrazine, benzidine,sulfide, and hydrogen gas (Stucki et al., 2002). Studies of non-tronites heated to 300 1C have shown partially irreversible forma-tion of Fe2+ species that induced an upward slope in reflectancespectra from ∼1–1.8 mm (Morris et al., 2009; R. V. Morris, pers.comm. 2011). As Fe3+ in nontronite is not expected to reducesimply by heating, a more complex mechanism likely transpired,possibly including organic admixtures in the sample. These Fe2+

species could be in the octahedral sites as postulated by Stucki(1988, 2006) through nontronite reduction reactions or as inter-layer Fe hydroxides as suggested by Meunier et al. (2010).

Table 2Biosignature preservation potential.

Highly favorable conditions for preservation of biosignaturesRapid burial in fine-grained clay-rich system FLow permeability environments with consistent low temperatures SStable minerals with long crustal residence times B

Less favorable conditions for preservation of biosignaturesBurial diagenesis and elevated temperatures DMinerals that readily react with external fluids FWeathering, erosion and alteration S

Minerals listed from highest to lowest preservation ability (Knoll, 1985; ButPhosphate, hematite, silica, phyllosilicate, carbonate, sulfate, halite

4.4. Implications of phyllosilicate-bearing rocks for the preservationof pre-biotic chemistry and/or biosignatures

Numerous models have been proposed for the origin of life onEarth. Some ideas considered include emergence of life associatedwith impact craters (Cockell, 2006) and hydrothermal solutions(e.g. Russell and Hall, 1997; Martin et al., 2008). In any of thesemodels Fe-rich smectites could have been present, which may berelated to chemical reactions essential to the origin of life (Harder,1986). Some of the oldest terrestrial rocks from the Hadean periodwere mafic (O'Neil et al., 2008) and thus Fe-rich smectites wouldhave been likely alteration products, although they are no longerpresent in any of the oldest terrestrial rocks found to date (Veldeand Meunier, 2008; Meunier et al., 2010). Thus, studying theancient, preserved Fe/Mg-smectites on Mars may provide insightsinto clay formation processes on the early Earth as well.

Preservation of biosignatures is favored in rapid burial condi-tions in fine-grained clay-rich systems (Farmer and Des Marais,1999). Long-term preservation is most successful in host rockscomposed of stable minerals that are resistant to weathering andprovide an impermeable barrier preventing mixing with externalfluids that would alter and/or remove the biosignatures. Mineralprecipitates such as phyllosilicates and silica provide an excellentmatrix for microbial fossilization (Farmer and Des Marais, 1999).Phyllosilicates have a longer crustal residence time than manyother minerals such as carbonates and sulfates, thus enablingpreservation of any biosignatures present (e.g. Butterfield, 1990;Summons et al., 2011). The preservation potential is highest forsediments formed in low permeability environments where tem-peratures remained low over time (Summons et al., 2011). Theconditions and minerals supporting preservation of biosignaturesare summarized in Table 2.

Possible biosignatures include cell-shaped objects (microbial fos-sils), remnants of biomolecules or microorganisms in fluid inclusions,the presence of polycyclic aromatic hydrocarbons (PAHs), and biogenicmineral structures and/or compositions (Des Marais and Walter, 1999;Farmer and DesMarais, 1999). Burial diagenesis is onemechanism thatdisrupts preservation of biosignatures. The absence of illitization ofsmectites at Mawrth Vallis indicates that burial diagenesis was notprevalent, thus increasing the chance of preservation of any biosigna-tures in this region. The Fe/Mg-smectite unit at Mawrth Vallis is likelyto be favored for containing organics and thus possibly microbes asFe-bearing smectites have been found to be more reactive withorganics. Advantages of this unit for biosignature preservation includeits thickness and homogeneity, compared to the upper Al/Si-rich unit,while the reactive nature of the Fe in smectites could have reduced thestability of biosignatures. Al-smectites and hydrated silica are alsosupportive environments for preservation of biosignatures, howeverkaolinite/halloysite and amorphous Al/Si phases, plus possible acid-treated materials suggests more active chemistry in this unit including

armer and Des Marais (1999)ummons et al. (2011)utterfield (1990) and Zhang et al. (2007)

es Marais and Walter (1999) and Farmer and Des Marais (1999)armer and Des Marais (1999)ummons et al. (2011)

terfield, 1990; Farmer and Des Marais, 1999; Summons et al., 2011)


leaching and high drainage and potentially less stable conditions forpreservation of biosignatures. The Fe2+ zone at the border of theFe/Mg-smectites and Al/Si-rich unit could be the best place to searchfor biosignatures, as Fe hydroxide complexes associated with thenontronite may have been reduced through contact with organics ormicrobes.

5. Summary and conclusions

The thick and widespread phyllosilicate outcrops observedin the Mawrth Vallis region indicate that abundant water waspresent here during the Noachian period. Factors shaping theformation and alteration of the observed phyllosilicates includeaqueous processes, chemistry, and perhaps biology. The stratigra-phy of Fe3+/Mg-smectite plus ferric oxide-bearing phases in thelower unit, covered by an Fe2+-bearing phyllosilicate, and thentopped by a unit containing ferric oxide-bearing phases plusAl-smectite, kaolin-group minerals, zeolite, amorphous Al/Siphases and/or acid-treated smectite, Fe–OH-bearing sulfate, orgibbsite is best explained by multiple aqueous events. The forma-tion of the Fe3+/Mg-smectite plus ferric oxide-bearing phaseslikely occurred in a neutral to slightly basic environment, probablyunder long-term aqueous conditions in order to achieve the100–200 m thick deposit. The Fe2+ and Al/Si-rich units could beexplained by an acid-leaching procedure where Fe/Mg are leachedout of a precursor phase. Also possible is a scenario following theChevrier model where Fe2+ clays would precipitate out first underlower pH conditions and species higher in Al and Si would crystal-lize later. These alteration episodes occurred at or close to thesurface, and the rocks have not been modified since these events bydeep burial or high water activity. Regardless of the process thatformed them, phyllosilicates have been detected across the planetin numerous sites where ancient crust is visible on the surface,suggesting that the processes forming phyllosilicates may havebeen widespread on early Mars and that only a fraction of thephyllosilicate-rich rocks are currently exposed on the surface.

Fe-rich smectites were one of the earliest types of phyllosili-cates present on the Earth and Mars. The thick sequence atMawrth Vallis likely formed from basaltic rocks under highwater/rock ratio conditions. Phyllosilicate-bearing rocks may havebeen an ideal place on Mars for pre-biotic chemistry and possiblythe development of life as well. Phyllosilicates, especially smec-tites, can serve as reaction surfaces that bind molecules andcatalyze chemical reactions. Numerous chemical reactions haveproduced RNA molecules from precursors on smectite surfaces.Other experiments have shown excellent survival of microbesin clay environments under extreme Mars-like temperature andhumidity conditions. If microbes were present on Mars, theancient Fe-smectite-bearing rocks could have been a favorableenvironment for them to evolve and possibly thrive. The expansiveFe/Mg-smectite-rich unit at Mawrth Vallis could have been anideal location for past life because the vast smectite outcropimplies persistent surface or subsurface water was once present.Preservation of any biosignatures left behind would be favoredin fine-grained smectite and silica-rich systems formed in lowpermeability environments where temperatures remained lowover time. As there has been little to no conversion of the smectiteat Mawrth Vallis to higher temperature phyllosilicates such asbeidellite, illite or chlorite, this region appears to have escapedburial diagenesis that likely occurred elsewhere on Mars. TheMawrth Vallis phyllosilicate outcrops are colored by changes inphyllosilicate chemistry that could indicate an active chemicalenvironment supportive of habitable conditions, that may haveeven provided energy sources for microbes, and that could in partbe a result of the presence of microbes. The results of this study

support sending a rover to Mawrth Vallis equipped with instru-ments and experiments able to search for biosignatures.

Acknowledgments

The authors thank the MRO CRISM and HiRISE science andoperations teams for acquiring the data and providing support toJ. Bishop for this work. Thanks are also due to NASA for supportingBrown University's RELAB facility. The authors are grateful to JavierCuadros and an anonymous reviewer for helpful comments.Support from a research fellow grant from the European SpaceAgency to D. Loizeau and from the National Science Foundationand the NASA Astrobiology Institute to L. Saper through theResearch Experience for Undergraduates program at the SETIInstitute are greatly appreciated.

References

Alt, J.C., Mata, P., 2000. On the role of microbes in the alteration of submarinebasaltic glass: a TEM study. Earth and Planetary Science Letters 181, 301–313.

Andrieux, P., Petit, S., 2010. Hydrothermal synthesis of dioctahedral smectites: theAl–Fe3+ chemical series: Part I: influence of experimental conditions. AppliedClay Science 48, 5–17.

Anders, E., 1989. Pre-biotic organic matter from comets and asteroids. Nature 342,255–257.

Anderson, J.H., Wickersheim, K.A., 1964. Near infrared characterization of water andhydroxyl groups on silica surfaces. Surface Science 2, 252–260.

Anderson, W.I., Stucki, J.W., 1979. Effect of structural Fe2+ on visible absorptionspectra of nontronite suspensions. In: Mortland, M.M., Farmer, V.C. (Eds.),Developments in Sedimentology. Proceedings of the VI International ClayConference, vol. 27. Elsevier, pp. 75–83.

Arrhenius, G., 1986. Dysoxic environments as models for primordial mineralisation.In: Cairns-Smith, A.G., Hartman, H. (Eds.), Clay Minerals and the Origin of LifeCambridge University Press, pp. 97–104.

Badaut, D., Besson, G., Decarreau, A., Rautureau, R., 1985. Occurrence of a ferroustrioctahedral smectite in reecnt sediments of Atlantis II Deep Red Sea. ClayMinerals 20, 389–404.

Bailey, S.W., 1980. Structures of layer silicates. In: Brindley, G.W., Brown, G. (Eds.),Crystal Structures of Clay Minerals and Their X-ray Identification, MineralogicalSociety Monograph Series, vol. 5. Mineralogical Society, pp. 1–124.

Bandfield, J.L., Rogers, D., Smith, M.D., Christensen, P.R., 2004. Atmosphericcorrection and surface spectral unit mapping using thermal emission imagingsystem data. Journal of Geophysical Research, http://dx.doi.org/10.1029/2004JE002289.

Banin, A., Rishpon, J., 1979. Smectite clays in Mars soil: evidence for their presenceand role in Viking biology experimental results. Journal of Molecular Evolution14, 133–152.

Banin, A., Margulies, L., 1983. Simulation of Viking biology experiments suggestssmectites not palagonites, as martian soil analogues. Nature 305, 523–525.

Banin, A., Han, F.X., Kan, I., Cicelsky, A., 1997. Acidic volatiles and the Mars soil.Journal of Geophysical Research 102, 13,341–13,356.

Benzerara, K., Yoon, T.H., Menguy, N., Tyliszczak, T., Brown, G.E., 2005. Nanoscaleenvironments associated with bioweathering of a Mg–Fe–pyroxene. Proceed-ings of the National Academy of Sciences of the United States of America 102,979–982.

Bergaya, F., Aouad, A., Mandalia, T., 2006. Pillared clays and clay minerals. In:Bergaya, F., Theng, B.K.G., Lagaly, G. (Eds.), Handbook of Clay Science, Devel-opments in Clay Science, vol. 1. Elsevier, pp. 393–421.

Bischoff, J.L., 1972. A ferroan nontronite from the Red Sea geothermal system. Claysand Clay Minerals 20, 217–223.

Bishop, J.L., Murad, E., 2002. Spectroscopic and Geochemical Analyses of Ferrihy-drite from Hydrothermal Springs in Iceland and Applications to Mars. In:Smellie, J.L., Chapman, M.G. (Eds.), Volcano-Ice Interactions on Earth and Mars.Geological Society, pp. 357–370, Special Publication no. 202.

Bishop, J.L., Pieters, C.M., Edwards, J.O., 1994. Infrared spectroscopic analyses on thenature of water in montmorillonite. Clays and Clay Minerals 42, 701–715.

Bishop, J.L., Murad, E., Dyar, M.D., 2002a. The influence of octahedral andtetrahedral cation substitution on the structure of smectites and serpentinesas observed through infrared spectroscopy. Clay Minerals 37, 617–628.

Bishop, J.L., Madeová, J., Komadel, P., Fröschl, H., 2002b. The influence of structuralFe, Al and Mg on the infrared OH bands in spectra of dioctahedral smectites.Clay Minerals 37, 607–616.

Bishop, J.L., Schiffman, P., Lane, M.D., Dyar, M.D., 2005. Solfataric alteration inHawaii as a mechanism for formation of the sulfates observed on Mars byOMEGA and the MER instruments. Lunar and Planetary Science XXXVI, Abstract#1456.

Bishop, J.L., Murad, E., 2005. The visible and infrared spectral properties of jarositeand alunite. American Miner. 90, 1100–1107.

http://refhub.elsevier.com/S0032-0633(13)00114-1/sbref1






















http://dx.doi.org/10.1029/2004JE002289

http://dx.doi.org/10.1029/2004JE002289

http://dx.doi.org/10.1029/2004JE002289

http://dx.doi.org/10.1029/2004JE002289




































Bishop, J.L., Lane, M.D., Dyar, M.D., Brown, A.J., 2008a. Reflectance and emissionspectroscopy study of four groups of phyllosilicates: smectites, kaolinite-serpentines, chlorites and micas. Clay Minerals 43, 35–54.

Bishop, J.L., Noe Dobrea, E.Z., McKeown, N.K., Parente, M., Ehlmann, B.L., Michalski,J.R., Milliken, R.E., Poulet, F., Swayze, G.A., Mustard, J.F., Murchie, S.L., Bibring,J.-P., 2008b. Phyllosilicate diversity and past aqueous activity revealed atMawrth Vallis, Mars. Science 321, 830–833, http://dx.doi.org/10.1126/science.1159699.

Bishop, J.L., Parente, M., Weitz, C.M., Noe Dobrea, E.Z., Roach, L.A., Murchie, S.L.,McGuire, P.C., McKeown, N.K., Rossi, C.M., Brown, A.J., Calvin, W.M., Milliken, R.E.,Mustard, J.F., 2009. Mineralogy of Juventae Chasma: sulfates in the Light-tonedMounds, Mafic Minerals in the Bedrock, and Hydrated Silica and HydroxylatedFerric Sulfate on the Plateau. Journal of Geophysical Research, 114, http://dx.doi.org/10.1029/2009JE003352.

Bishop, J.L., Gates, W.P., Makarewicz, H.D., McKeown, N.K., Hiroi, T., 2011a.Reflectance spectroscopy of beidellites and their importance for Mars. Claysand Clay Minerals 59, 376–397.

Bishop, J.L., Saper, L., Beyer, R.A., Lowe, D.R., Wray, J.J., McKeown, N.K., Parente, M.,2011b. Possible sedimentary features in phyllosilicate-bearing rocks at MawrthVallis, Mars, Proceedings of the 42nd Lunar Planetary Science Conference, abs#2374.

Bishop, J.L., Rampe, E.B., Bish, D.L., Baker, L.L., Abidin, Z., Matsue, N., Henmi, T., 2013.Spectral and hydration properties of allophane and imogolite. Clays and ClayMinerals 61, 57–74.

Boles, J.R., 1971. Synthesis of analcime from natural heulandite and clinoptillolite.American Mineralogist 56, 1724–1734.

Butterfield, N.J., 1990. Organic preservation of non-mineralizing organisms and thetaphonomy of the Burgess Shale. Paleobiology 16, 272–286.

Caillaud, J., Proust, D., Right, D., 2006. Weathering sequences of rock-formingminerals in a serpentinite: influence of microsystems on clay mineralogy. Claysand Clay Minerals 54, 87–100.

Carter, J., Poulet, F., Bibring, J.-P., Murchie, S., 2010. Detection of hydrated silicates incrustal outcrops in the Northern Plains of Mars. Science 328, 1682–1686.

Carter, J., Poulet, F., Ody, A., Bibring, J.P., Murchie, S., 2011. Global distribution,composition and setting of hydrous minerals on Mars: a reappraisal, Proceed-ings of the 42nd Lunar and Planetary Science Conference, abstract #1608.

Chamley, H., 1989. Clay Sedimentology. Springer-Verlag.Chevrier, V., Poulet, F., Bibring, J.-P., 2007. Early geochemical environment of

Mars as determined from thermodynamics of phyllosilicates. Nature 448,60–63.

Chipera, S.J., Bish, D.L., 1999. Thermodynamic modeling of natural zeolite stability.In: Kodama, H., Mermut, A.R., Torrance, J.K. (Eds.), Proceedings of the 11thInternational Clay Conference, June, 1997, pp. 595–602.

Chyba, C.F., Thomas, P.J., Brookshaw, L., Sagan, C., 1990. Cometary delivery oforganic molecules to the early Earth. Science 249, 366–373.

Clark, R.N., Swayze, G.A., Wise, R., Livo, E., Hoefen, T., Kokaly, R., Sutley, S.J., 2007.USGS digital spectral library splib06a. In Digital Data Series 231. U.S. GeologicalSurvey.

Cockell, C.S., 2006. The origin and emergence of life under impact bombardment.Philosophical Transactions of the Royal Society B: Biological Sciences 361,1845–1856.

Coombs, D.S., Bosel, C.A., Kawachi, Y., Paterson, L.A., 2005. A silica-deficient,shallow-marine zeolite assemblage in the Foveaux Formation, Bluff Peninsula,New Zealand. Mineralogical Magazine 69, 137–144.

Cotton, A., 2008. Dissolution kinetics of clinoptilolite and heulandite in alkalineconditions. Bioscience Horizons 1, 38–43.

Cuadros, J., Dekov, V.M., Arroyo, X., Nieto, F., 2011. Smectite formation in submarinehydrothermal sediments: samples fromtheHMSchallenger expedition (1872–1876).Clays and Clay Minerals 59, 147–164.

Dashman, T., Stotzky, T., 1982. Adsorption and binding of aminoacids on homoionicmontmorillonite and kaolinite. Soil Biology and Biochemistry 14, 447–456.

Decarreau, A., Petit, S., Martin, F., Vieillard, P., Joussein, E., 2008. Hydrothermalsynthesis, between 75 and 150 C, of high-charge ferric nontronites. Clays andClay Mineral 56, 322–337.

Deer, W.A., Howie, R.A., Zussman, J., 1992. An Introduction to the Rock-FormingMinerals. Longman.

Delsemme, A.H., 1997. The origin of the atmosphere and of the oceans. In: Thomas, P.J.,Chyba, C.F., McKay, C.P. (Eds.), Comets and the Origin and Evolution of Life.Springer-Verlag, pp. 29–67.

Des Marais, D.J., Walter, M.R., 1999. Astrobiology: exploring the origins, evolution,and distribution of life in the universe. Annual Review of Ecology andSystematics 30, 397–420.

Desprairies, A., Tremblay, P., Laloy, C., 1989. Secondary mineral assemblages in avolcanic sequence drilled during ODP Leg 104 in the Norwegian Sea. Proceed-ings of the Ocean Drilling Program-Scientific Results, 397–409.

Dong, H., Kostka, J.E., Kim, J.W., 2003. Microscopic evidence for microbial dissolu-tion of smectite. Clays and Clay Minerals 51, 502–512.

Dong, H., Jaisi, D.P., Km, J., Zhang, G., 2009. Microbe-clay mineral interactions.American Mineralogist 94, 1505–1519.

Drief, A., Martinez Ruiz, F., Nieto, F., Velilla Sanchez, N., 2002. Transmissionelectron microscopy evidence for experimental illitization of smectite inK-enriched seewater solution at 50 C and basic pH. Clays and Clay Minerals50, 746–756.

Drodt, M., Trautwein, A.X., König, I., Suess, E., Bender Koch, C., 1997. Mössbauerspectroscopic studies on the iron forms of deep-sea sediments. Physics andChemistry of Minerals 24, 281–293.

Drodt, M., Lougear, A., Trautwein, A.X., König, I., Suess, E., Bender Koch, C., 1998.Studies of iron in deep-sea sediments by Mössbauer spectroscopy. HyperfineInteractions 117, 383–403.

Dyar, M.D., Gunter, M.E., 2007. Mineralogy and Optical Mineralogy. MineralogicalSociety.

Ernstsen, V., Gates, W.P., Stucki, J.W., 1998. Microbial reduction of structural iron inclays—A renewable source of reduction capacity. Journal of EnvironmentalQuality 27, 761–766.

Ehlmann, B.L., Mustard, J.F., Fasset, C.I., Schon, S.C., Head, J.W., Des Marais, D.J.,Grant, J.A., Murchie, S.L., 2008a. Clay minerals in delta deposits and organicpreservation potential on Mars. Nature Geoscience 1, 355–358, http://dx.doi.org/10.1038/ngeo207.

Ehlmann, B.L., Mustard, J.F., Murchie, S.L., Poulet, F., Bishop, J.L., Brown, A.J., Calvin, W.M.,Clark, R.N., Des Marais, D.J., Milliken, R.E., Roach, L.H., Roush, T.L., Swayze, G.A., Wray,J.J., 2008b. Oribtal identification of carbonate-bearing rocks on Mars. Science 322,1828–1832.

Ehlmann, B.L., Mustard, J.F., Swayze, G.A., Clark, R.N., Bishop, J.L., Poulet, F., Marais,D.J.D., Roach, L.H., Milliken, R.E., Wray, J.J., Barnouin-Jha, O., Murchie, S.L., 2009.Identification of hydrated silicate minerals on Mars using MRO-CRISM: geologiccontext near Nili Fossae and implications for aqueous alteration. Journal ofGeophysical Research, 114, http://dx.doi.org/10.1029/2009JE003339.

Ehlmann, B.L., Mustard, J.F., Murchie, S.L., Bibring, J.P., Meunier, A., Fraeman, A.A.,Langevin, Y., 2011. Subsurface water and clay mineral formation during theearly history of Mars. Nature 479, 53–60.

Farmer, V.C., Russell, J.D., 1971. Interlayer complexes in layer silicates: the structureof water in lamellar ionic solutions. Transactions of the Faraday Society 67,2737–2749.

Farmer, V.C., McHardy, W.J., Elsass, F., Robert, M., 1994. hk-Ordering in aluminousnontronite and saponite synthesized near 90 1C: effects of synthesis conditionsonnontronite composition and ordering. Clays and Clay Minerals 42, 180–186.

Farmer, J.D., Des Marais, D.J., 1999. Exploring for a record of ancient Martian life.Journal of Geophysical Research 104, 26,977–26,995.

Faye, G.H., 1968. The optical absorption spectra of iron in six-coordinate sites inchlorite, biotite, phlogopite and vivianite; some aspects of pleochroism in thesheet silicates. Canadian Mineralogist 9, 403–425.

Ferris, J.P., 2005. Mineral catalysis and prebiotic synthesis: Montmorillonite-catalyzed formation of RNA. Elements 1, 145–149.

Ferris, J.P., 2006. Montmorillonite-catalysed formation of RNA oligomers: thepossible role of catalysis in the origins of life. Philosophical Transactions ofthe Royal Society B: Biological Sciences 361, 1777–1786.

Ferris, J.P., Hagan Jr., W.J., 1986. The adsorption and reaction of adenine nucleotideson montmorillonite. Origins of Life and Evolution of the Biosphere 17, 69–84.

Ferris, J.P., Huang, C.-H., Hagan, W.J., 1988. Montmorillonite: a multifunctionalmineral catalyst for the prebiological formation of phosphate esters. Origins ofLife and Evolution of Biospheres 18, 121–133.

Ferris, J.P., Ertem, G., Agarwal, V.K., 1989. The adsorption of nucleotides andpolynucleotides of montmorillonite clay. Origins of Life and Evolution of theBiosphere 19, 153–164.

Fialips, C.-I., Huo, D., Yan, L., Wu, J., Stucki, J.W., 2002. Effect of Fe oxidation state onthe IR spectra of Garfield nontronite. American Mineralogist 87, 630–641.

Fisk, M.R., Giovannoni, S.J., Thorseth, I.H., 1998. Alteration of oceanic volcanic glass:textural evidence of microbial activity. Science 281, 978–980.

Foster, T.L., Winans Jr., L., Casey, R.C., Kirschner, L.E., 1978. Response of terrestrialmicroorganisms to a simulated Martian environment. Applied and Environ-mental Microbiology 35, 730–737.

Franchi, M., Ferris, J.P., Gallori, E., 2003. Cations as mediators of the adsorption ofnucleic acids on clay surfaces in prebiotic environments. Origins of Life andEvolution of the Biosphere 33, 1–16.

Fripiat, J.J., Chaussidon, J., Touillaux, R., 1960. Study of dehydration of montmor-illonite and vermiculite by infrared spectroscopy. Journal of Physical Chemistry64, 1234–1241.

Furukawa, Y., O'Reilly, S.E., 2007. Rapid precipitation of amorphous silica inexperimental systems with nontronite (NAu-1) and Shewanella oneidensisMR-1. Geochimica et Cosmochimica Acta 71, 363–377.

Gates, W.P., Jaunet, A.M., Tessier, D., Cole, M.A., Wilkinson, H.T., Stucki, J.W., 1998.Swelling and texture of iron-bearing smectites reduced by bacteria. Clays andClay Minerals 46, 487–497.

Gooding, J.L., 1978. Chemical weathering on Mars. Thermodynamic stabilities ofprimary minerals (and their alteration products) from mafic igneous rocks.Icarus 33, 483–513.

Grauby, O., Petit, S., Decarreau, A., Baronnet, A., 1994. The nontronite-saponiteseries: an experimental approach. European Journal of Mineralogy 6, 99–112.

Güven, N., 1988. Smectites. In Reviews in Mineralogy. In: Bailey, S.W. (Ed.), HydrousPhyllosilicates, vol. 19. Mineralogical Soc. America, pp. 497–559.

Guthrie, G.D., Bish, D.L., Reynolds, R.L., 1995. Modeling the X-ray diffraction patternof opal-CT. American Mineralogist 80, 869–872.

Hall, A., 1998. Zeolitization of volcaniclastic sediments; the role of temperature andpH. Journal of Sedimentary Research 68, 739–745.

Harder, H., 1976. Nontronite synthesis at low temperatures. Chemical Geology 18,169–180.

Harder, H., 1988. Synthesis of iron-rich clays in environments with little or nooxygen. In: Cairns-Smith, A.G., Hartman, H. (Eds.), Clay Minerals and The Originof Life. Cambridge University Press, pp. 91–96.

Harrison, K.P., Grimm, R.E., 2005. Groundwater-controlled valley networks and thedecline of surface runoff on early Mars. Journal of Geophysical Research 110,E12S16, http://dx.doi.org/10.1029/2005JE002455.




http://dx.doi.org/10.1126/science.1159699




http://dx.doi.org/10.1029/2009JE003352

http://dx.doi.org/10.1029/2009JE003352

http://dx.doi.org/10.1029/2009JE003352

http://dx.doi.org/10.1029/2009JE003352
































http://refhub.elsevier.com/S0032-0633(13)00114-1/othref0005



















































http://dx.doi.org/10.1038/ngeo207








http://dx.doi.org/10.1029/2009JE003339

http://dx.doi.org/10.1029/2009JE003339

http://dx.doi.org/10.1029/2009JE003339
































































http://dx.doi.org/10.1029/2005JE002455

http://dx.doi.org/10.1029/2005JE002455

http://dx.doi.org/10.1029/2005JE002455


Hawrylewicz, E., Gowdy, B., Ehrlich, R., 1962. Micro-organisms under a simulatedMartian environment. Nature 193, 497.

Hedges, J.I., Hare, P.E., 1987. Amino acid adsorption by clay minerals in distilledwater. Geochimica et Cosmochimica Acta 51, 255–259.

Hower, J., Eslinger, E.V., Hower, M.E., Perry, E.A., 1976. Mechanism of burialmetamorphism of argillaceous sediments. 1. Mineralogical and chemicalevidence. Geological Society of America Bulletin 87, 725–737.

Huggett, J.M., Cuadros, J., 2005. Low-temperature illitization of smectite in the lateeocene and early oligocene of the Isle of Wight (Hampshire basin), U.K.American Mineralogist 90, 1192–1202.

Huggett, J.M., Cuadros, J., 2010. Glauconite formation in lacustrine/palaeosolsediments, Isle of Wight (Hampshire Basin), UK. Clay Minerals 45, 35–49.

Jackson, M.L., 1959. Frequency distribution of clay minerals in major great soilgroups as related to the factors of soil formation. Clays and Clay Minerals 6,133–143.

Jaisi, D.P., Dong, H., Liu, C., 2007. Influence of biogenic Fe(II) on the extent ofmicrobial reduction of Fe(III) in clay mineral nontronite, illite, and chlorite.Geochimica et Cosmochimica Acta 71, 1145–1158.

Kanavarioti, A., Mancinelli, R.L., 1990. Could organic matter have been preserved onMars for 3.5 billion years? Icarus 84, 196–202.

Kawano, M., Tomita, K., 1997. Experimental study on the formation of zeolites fromobsidian by interaction with NaOH and KOH solutions at 150 and 200 1C. Claysand Clay Minerals 45, 365–377.

Kikuchi, E., Matsuda, T., 1988. Shape selective acid catalysis by pillared clays.Catalysis Today 2, 297–307.

Kim, J.W., Dong, H., Seabaugh, J., Newell, S.W., Eberl, D.D., 2004. Role of microbes inthe smectite-to-illite reaction. Science 303, 830–832.

Knoll, A.H., 1985. Exceptional preservation of photosynthetic organisms in silicifiedcarbonates and silicified peats. Philosophical Transactions of the Royal SocietyB: Biological Sciences 311, 111–122.

König, I., Drodt, M., Suess, E., Trautwein, A.X., 1997. Iron-reduction through the tan-green color transition in deep-sea sediments. Geochimica et CosmochimicaActa 61, 1679–1683.

König, I., Haeckel, M., Drodt, M., Suess, E., Trautwein, A.X., 1999. Reactive Fe(II)layers in deep-sea sediments. Geochimica et Cosmochimica Acta 63, 1517–1526.

Konhauser, K.O., Schiffman, P., Fisher, Q.J., 2002. Microbial mediation of authigenicclays during hydrothermal alteration of basaltic tephra, Kilauea Volcano.Geochemistry Geophysics Geosystems 3, 1075, http://dx.doi.org/10.1029/2002GC000317.

Lambert, J.-F., 2008. Adsorption and polymerization of amino acids on mineralsurfaces: a review. Origins of Life and Evolution of Biospheres 28, 211–242.

Lawless, J.G., 1986. Clay-organic interactions and the origiin of life. In: Cairns-Smith, A.G.,Hartman, H. (Eds.), Clay Minerals and the Origin of Life. Cambridge University Press,pp. 135–137.

Lawless, J.G., Banin, A., Church, F.M., Mazzurco, J., Huff, J., Kao, R., Cook, A., Lowe, T.,Orenberg, J.B., Edelson, E., 1985. pH profile of the adsorption of nucleotidesonto montmorillonite. I. Selected homoionic clays. Origins of Life and Evolutionof the Biosphere 15, 77–88.

Lazcano-Araujo, A.R., Oro, J., 1981. Cometary material and the origins of life onEarth. In: Ponnamperuma, C. (Ed.), Comets and the Origin of Life D. ReidelPublishing Company, pp. 191–225.

Lear, P.R., Stucki, J.W., 1987. Intervalence electron transfer and magnetic exchangein reduced nontronite. Clays and Clay Minerals 35, 373–378.

Lee, K., Kostka, J.E., Stucki, J.W., 2006. Comparisons of structural Fe reduction insmectites by bacteria and dithionite: an infrared spectroscopic study. Clays andClay Minerals 54, 195–208.

Loizeau, D., Mangold, N., Poulet, F., Ansan, V., Hauber, E., Bibring, J.P., Gondet, B.,Langevin, Y., Masson, P., Neukum, G., 2010. Stratigraphy in the Mawrth Vallisregion through OMEGA, HRSC color imagery and DTM. Icarus 205, 396–418.

Loizeau, D., Mangold, N., Poulet, F., Bibring, J.-P., Gendrin, A., Ansan, V., Gomez, C.,Gondet, B., Langevin, Y., Masson, P., Neukum, G., 2007. Phyllosilicates in theMawrth Vallis region of Mars. Journal of Geophysical Research, 112, http://dx.doi.org/10.1029/2006JE002877.

Lonsdale, P.F., Bischoff, J.L., Burns, V.M., Kastner, M., Sweeney, R.E., 1980. A high-temperature hydrothermal deposit on the seabed at a Gulf of Californiaspreading center. Earth and Planetary Science Letters 49, 8–20.

Lyle, M., 1983. The brown-green color transition in marine sediments: a marker ofthe Fe(III)–Fe(II) redox boundary. Limnology and Oceanography 28, 1026–1033.

Madejová, J., Pentrák, M., Pálková, H., Komadel, P., 2009. Near-infrared spectro-scopy: a powerful tool in studies of acid-treated clay minerals. VibrationalSpectroscopy 49, 211–218.

Manceau, A., Drits, V.A., Lanson, B., Chateigner, D., Wu, J., Huo, D., Gates, W.P.,Stucki, J.W., 2000. Oxidation-reduction mechanism of iron in dioctahedralsmectites: II. Crystal chemistry of reduced Garfield nontronite. AmericanMineralogist 85, 153–172.

Martin, W., Baross, J., Kelley, D., Russell, M.J., 2008. Hydrothermal vents and theorigin of life. Nature Reviews Microbiology 6, 805–814.

McKeown, N.K., 2010. Water on Ancient Mars: Forming the Phyllosilicates atMawrth Vallis. University of California at Santa Cruz.

McKeown, N.K., Bishop, J.L., Cuadros, J., Hillier, S., Amador, E., Makarewicz, H.D.,Parente, M., Silver, E., 2011. Interpretation of reflectance spectra of claymineral–silica mixtures: implications for Martian clay mineralogy at MawrthVallis. Clays and Clay Mineral 59, 400–415.

McKeown, N.K., Bishop, J.L., Noe Dobrea, E.Z., Ehlmann, B.L., Parente, M., Mustard, J.F.,Murchie, S.L., Swayze, G.A., Bibring, J.-P., Silver, E., 2009. Characterization ofphyllosilicates observed in the central Mawrth Vallis region, Mars, their potential

formational processes, and implications for past climate. Journal of GeophysicalResearch, 114, http://dx.doi.org/10.1029/2008JE003301.

McLennan, S.M., 2003. Sedimentary silica on Mars. Geology 31, 315–318.McSween, H.Y.J., Taylor, G.J., Wyat, M.B., 2009. Elemental composition of the

Martian crust. Science 324, 736–739.Meunier, A., 2007. Soil hydroxy-interlayered minerals: a re-interpretation of their

crystallochemical properties. Clays and Clay Minerals 55, 380–388.Meunier, A., Petit, S., Cockell, C.S., Albani, A.E., Beaufort, D., 2010. The Fe-rich clay

microsystems in basalt-komatiite lavas: importance of Fe-smectites for pre-biotic molecule catalysis during the Hadean Eon. Origins of Life and Evolutionof the Biosphere 40, 253–272.

Michalski, J.R., Noe Dobrea, E.Z., 2007. Evidence for a sedimentary origin of clayminerals in the Mawrth Vallis region, Mars. Geology 35, 951–954.

Michalski, J.R., Kraft, M.D., Sharp, T.G., Williams, L.B., Christensen, P.R., 2006.Emission spectroscopy of clay minerals and evidence for poorly crystallinealuminosilicates on Mars from Thermal Emission Spectrometer data. Journalof Geophysical Research 111, E03004, http://dx.doi.org/10.1029/2005JE002438.

Michalski, J.R., Poulet, F., Loizeau, D., Mangold, N., Noe Dobrea, E.Z., Bishop, J.L., Wray, J.J.,McKeown, N.K., Parente, M., Hauber, E., Altieri, F., Carrozzo, F.G., Niles, P.B., 2010. TheMawrth Vallis region of Mars: a potential landing site for the Mars ScienceLaboratory (MSL) mission. Astrobiology 10, 687–703.

Milliken, R.E., Swayze, G.A., Arvidson, R.E., Bishop, J.L., Clark, R.N., Ehlmann, B.L.,Green, R.O., Grotzinger, J., Morris, R.V., Murchie, S.L., Mustard, J.F., Weitz, C.M.,2008. Opaline silica in young deposits on Mars. Geology 36, 847–850, http://dx.doi.org/10.1130/G24967A.1.

Moll, D.M., Vestal, J.R., 1992. Survival of microorganisms in smectite clays:implications for Martian exobiology. Icarus 98, 233–239.

Morris, R.V., Graff, T.G., Mertzman, S.A., Lane, M.D., Christensen, P.R., 2003.Palagonitic (Not Andesitic) Mars: evidence from Thermal Emission and VNIRSpectra of Palagonitic Alteration Rinds on Basaltic Rock. In: Sixth InternationalConference on Mars, CD-Rom #3211 (abstr).

Morris, R.V., Golden, D.C., Ming, D.W., Graff, T.G., Arvidson, R.E., Wiseman, S.M.,Lichtenberg, K.A., 2009. Visible and near-IR reflectance spectra for smectite,sulfate and perchlorate under dry conditions for interpretation of martiansurface mineralogy. In: Proceedings of the Lunar and Planetary ScienceConference, abs. #2317.

Murchie, S., Arvidson, R., Bedini, P., Beisser, K., Bibring, J.-P., Bishop, J., Boldt, J.,Cavender, P., Choo, T., Clancy, R.T., Darlington, E.H., Des Marais, D., Espiritu, R.,Fort, D., Green, R., Guinness, E., Hayes, J., Hash, C., Heffernan, K., Hemmler, J.,Heyler, G., Humm, D., Hutcheson, J., Izenberg, N., Lee, R., Lees, J., Lohr, D.,Malaret, E., Martin, T., McGovern, J.A., McGuire, P., Morris, R., Mustard, J.,Pelkey, S., Rhodes, E., Robinson, M., Roush, T., Schaefer, E., Seagrave, G.,Seelos, F., Silvergate, P., Slavney, S., Smith, M., Shyong, W.-J., Strohbehn, K.,Taylor, H., Thompson, P., Tossman, B., Wirzburger, M., Wolff, M., 2007. CompactReconnaissance Imaging Spectrometer for Mars (CRISM) on Mars Reconnais-sance Orbiter (MRO). Journal of Geophysical Research 112, http://dx.doi.org/10.1029/2006JE002682.

Murchie, S.L., Roach, L.H., Seelos, F.P., Milliken, R.E., Mustard, J.F., Arvidson, R.E.,Wiseman, S., Lichtenberg, K., Andrews-Hanna, J., Bibring, J.-P., Bishop, J.L.,Parente, M., Morris, R.V., 2009a. Evidence for the origin of layered depositsin Candor Chasma, Mars, from mineral composition and hydrologic modeling.Journal of Geophysical Research, 114, http://dx.doi.org/10.1029/2009JE003343.

Murchie, S.L., Mustard, J.F., Ehlmann, B.L., Milliken, R.E., Bishop, J.L., McKeown, N.K.,Noe Dobrea, E.Z., Seelos, F.P., Buczkowski, D.L., Wiseman, S.M., Arvidson, R.E.,Wray, J.J., Swayze, G.A., Clark, R.N., Des Marais, D.J., McEwen, A.S., Bibring, J.P.,2009b. A synthesis of Martian aqueous mineralogy after 1 Mars year ofobservations from the Mars Reconnaissance Orbiter. Journal of GeophysicalResearch 114, http://dx.doi.org/10.1029/2009JE003342.

Murchie, S.L., Seelos, F.P., Hash, C.D., Humm, D.C., Malaret, E., McGovern, J.A., Choo, T.H.,Seelos, K.D., Buczkowski, D.L., Morgan, M.F., Barnouin-Jha, O.S., Nair, H., Taylor, H.W.,Patterson, G.W., Harvel, C.A., Mustard, J.F., Arvidson, R.E., McGuire, P., Smith, M.D.,Wolff, M.J., Titus, T.N., Bibring, J.-P., Poulet, F., 2009c. The Compact ReconnaissanceImaging Spectrometer for Mars investigation and data set from the Mars Recon-naissance Orbiter's primary science phase. Journal of Geophysical Research 114,http://dx.doi.org/10.1029/2009JE003344.

Mustard, J.F., Murchie, S.L., Pelkey, S.M., Ehlmann, B.L., Milliken, R.E., Grant, J.A.,Bibring, J.-P., Poulet, F., Bishop, J.L., Noe Dobrea, E.Z., Roach, L.A., Seelos, F.,Arvidson, R.E., Wiseman, S., Green, R., Hash, C., Humm, D., Malaret, E.,McGovern, J.A., Seelos, K., Clancy, R.T., Clark, R.N., Des Marais, D., Izenberg, N.,Knudson, A.T., Langevin, Y., Martin, T., McGuire, P., Morris, R.V., Robinson, M.,Roush, T., Smith, M., Swayze, G.A., Taylor, H., Titus, T.N., Wolff, M., 2008.Hydrated silicate minerals on Mars observed by the Mars ReconnaissanceOrbiter CRISM instrument. Nature 454, 305–309, http://dx.doi.org/10.1038/nature07097.

Nagy, K.L., 1995. Dissolution and precipitation kinetics of sheet silicates. In: White, A.F.,Brantley, S.L. (Eds.), Chemical Weathering Rates of Silicate Minerals, vol. 31.Mineralogical Society of America, pp. 173–233.

Nealson, K., 1997. The limits of life on Earth and searching for life on Mars. Journalof Geophysical Research 102, 23675–23686.

Neumann, A., Petit, S., Hofstetter, T.B., 2011. Evaluation of redox-active iron sites insmectites using middle and near infrared spectroscopy. Geochimica et Cosmo-chimica Acta 75, 2336–2355.

Noe Dobrea, E.Z., Bishop, J.L., McKeown, N.K., Fu, R., Rossi, C.M., Michalski, J.R.,Heinlein, C., Hanus, V., Poulet, F., Arvidson, R., Mustard, J.F., Ehlmann, B.L.,Murchie, S., McEwen, A.S., Swayze, G., Bibring, J.-P., Malaret, J.F.E., Hash, C.,





































http://dx.doi.org/10.1029/2002GC000317

http://dx.doi.org/10.1029/2002GC000317

http://dx.doi.org/10.1029/2002GC000317

http://dx.doi.org/10.1029/2002GC000317





















http://dx.doi.org/10.1029/2006JE002877

http://dx.doi.org/10.1029/2006JE002877

http://dx.doi.org/10.1029/2006JE002877

http://dx.doi.org/10.1029/2006JE002877





















http://dx.doi.org/10.1029/2008JE003301

http://dx.doi.org/10.1029/2008JE003301

http://dx.doi.org/10.1029/2008JE003301












http://dx.doi.org/10.1029/2005JE002438

http://dx.doi.org/10.1029/2005JE002438

http://dx.doi.org/10.1029/2005JE002438

http://dx.doi.org/10.1029/2005JE002438





http://dx.doi.org/10.1130/G24967A.1















http://dx.doi.org/10.1029/2006JE002682

http://dx.doi.org/10.1029/2006JE002682

http://dx.doi.org/10.1029/2006JE002682

http://dx.doi.org/10.1029/2006JE002682

http://dx.doi.org/10.1029/2009JE003343

http://dx.doi.org/10.1029/2009JE003343

http://dx.doi.org/10.1029/2009JE003343

http://dx.doi.org/10.1029/2009JE003342

http://dx.doi.org/10.1029/2009JE003342

http://dx.doi.org/10.1029/2009JE003342

http://dx.doi.org/10.1029/2009JE003344

http://dx.doi.org/10.1029/2009JE003344

http://dx.doi.org/10.1029/2009JE003344

http://dx.doi.org/10.1038/nature07097













2010. Mineralogy and Stratigraphy of Phyllosilicate-bearing and dark mantlingunits in the greater Mawrth Vallis/west Arabia Terra area: constraints ongeological origin. Journal of Geophysical Research 115, http://dx.doi.org/10.1029/2009JE003351.

Odin, G.S., 1986. The origin of clays on the Earth. In: Cairns-Smith, A.G., Hartman, H.(Eds.), Clay Minerals and the Origin of Life. Cambridge University Press,pp. 81–89.

Odom, D.G., Rao, M., Lawless, J.G., Oro, J., 1979. Association of nucleotides withhomoionic clays. Journal of Molecular Evolution 12, 365–367.

O'Hanley, D.S., Dyar, M.D., 1993. The composition of lizardite 1T and the formationof magnetite in serpentines. American Miner. 78, 391–404.

O'Neil, J., Carlson, R.W., Francis, D., Stevenson, R.K., 2008. Neodymium-142 Evidencefor Hadean Mafic Crust. Science 321, 1828–11831.

Orgel, L.E., 1998. Polymerization on the rocks: theoretical introduction. Origins ofLife and Evolution of Biospheres 28, 227–234.

Orgel, L.E., 2004. Prebiotic Chemistry and the Origin of the RNA World. CriticalReviews in Biochemistry and Molecular Biology 39, 99–123.

Parente, M., 2008. A new approach to denoising CRISM images. In: Proceedings ofthe Lunar Planetary Science Conference, Abstract #2528.

Parthasarathy, G., Choudary, B.M., Sreedhar, B., Kunwar, A.C., Srinivasan, R., 2003.Ferroud saponite from the Deccan trap, India, and its application in adsorptionand reduction of hexavalent chromium. American Mineralogist 88, 1983–1988.

Pelkey, S.M., Mustard, J.F., Murchie, S., Clancy, R.T., Wolff, M., Smith, M., Milliken, R.E.,Bibring, J.-P., Gendrin, A., Poulet, F., Langevin, Y., Gondet, B., 2007. CRISMmultispectral summary products: parameterizing mineral diversity on Marsfrom reflectance. Journal of Geophysical Research 112, http://dx.doi.org/10.1029/2006JE002831.

Pierazzo, E., Chyba, C.F., 1999. Amino acid survival in large cometary impacts.Meteoritics and Planetary Science 34, 909–918.

Pieri, D., 1976. Distribution of small channels on the Martian surface. Icarus 27,25–50.

Pinnavaia, T.J., 1983. Intercalated clay catalysts. Science 220, 365.Pinnavaia, T.J., Tzou, M.-S., Landau, S.D., Raythatha, R.H., 1984. On the pillaring and

delamination of smectite clay catalysts by polyoxo cations of aluminum. Journalof Molecular Catalysis 27, 195–212.

Pinnavaia, T.J., Mortland, M.M., 1986. In: Cairns-Smith, A.G., Hartman, H. (Eds.), ClayMinerals and the Origin of Life. Cambridge University Press, pp. 131–135.

Pollack, J.B., Kasting, J.F., Richardson, S.M., Poliackoff, K., 1987. The case for a warm,wet climate on early Mars. Icarus 71, 203–224.

Post, J.L., Plummer, C.C., 1972. The chlorite series of Flagstaff Hill area, California: apreliminary investigation, 20. Clays and Clay Mineral pp. 271–283.

Poulet, F., Bibring, J.-P., Mustard, J.F., Gendrin, A., Mangold, N., Langevin, Y.,Arvidson, R.E., Gondet, B., Gomez, C., 2005. Phyllosilicates on Mars andimplications for early Martian climate. Nature 438, 623–627.

Poulet, F., Mangold, N., Loizeau, D., Bibring, J.P., Langevin, Y., Michalski, J.R., Gondet, B.,2008. Abundance of minerals in the phyllosilicate-rich units on Mars. Astronomyand Astrophysics 487, http://dx.doi.org/10.1051/0004-6361:200810150.

Rampe, E.B., Kraft, M.D., Sharp, T.G., 2011. Allophane at Mawrth Vallis: implicationsfor aqueous alteration history. Eos Trans. AGU 92, Fall Meeting Supplement,Abstract P31C-1716.

Ruff, S.W., 2004. Spectral evidence for zeolite in the dust on Mars. Icarus 168,131–143.

Russell, J.D., Farmer, V.C., 1964. Infra-red spectroscopic study of the dehydration ofmontmorillonite and saponite. Clay Minerals Bulletin 5, 443–464.

Russell, M.J., Hall, A.J., 1997. The emergence of life from iron monosulphide bubblesat a submarine hydrothermal redox and pH front. Journal of the GeologicalSociety 154, 377–402.

Sanz-Montero, M.E., Rodríguez-Aranda, J.P., Pérez-Soba, C., 2009. Microbial weath-ering of Fe-rich phyllosilicates and formation of pyrite in the dolomiteprecipitating environment of a Miocene lacustrine system. European Journalof Mineralogy 21, 163–175.

Saper, L. Bishop, J.L., 2011. Reflectance spectroscopy of nontronite and ripidolitemineral mixtures in context of phyllosilicate unit composition at Mawrth Vallis.In: Proceedings of the Lunar Planetary Science Conference IVII, abs #2029.

Seabaugh, J.L., Dong, H., Kukkadapu, R.K., Eberl, D.D., Morton, J.P., Kim, J., 2006.Microbial reduction of Fe(III) in the Fithian and Muloorina illites: contrastingextents and rates of bioreduction. Clays and Clay Minerals 54, 67–79.

Schultz, P.H., 1986. The martian atmosphere before and after the Argyre impact.MECA Workshop on the Evolution of the Martian Atmosphere, 38.

Seelos, F.P., 2011. CRISM data processing and analysis products update—calibration,correction, and visualization. In: Proceedings of the 42nd Lunar PlanetaryScience Conference, abs #1438.

Siever, R., 1962. Silica solubility, 0–200 1C, and the diagenesis of siliceous sedi-ments. The Journal of Geology 70, 127–150.

Squyres, S.W., 1989. Urey Prize lecture: water on Mars. Icarus 79, 229–288.Stucki, J.W., 1997. Redox processes in smectites: soil environmental significance. In:

Auerswald, H., Stanjek, H. (Eds.), Advances in GeoEcology, vol. 30. Catena-Verlag, pp. 395–406.

Stucki, J.W., 2006. Properties and behaviour of iron in clay minerals. In: Bergaya, F.,G.L., Theng, B.G.K. (Eds.), Handbook of Clay Science. Elsevier, pp. 423–476.

Stucki, J.W., Lee, K., Zhang, L., Larson, R.A., 2002. Effects of iron oxidation state onthe surface and structural properties of smectites. Pure Applied Chemistry 74,2145–2158.

Stucki, J.W., Komadel, P., Wilkinson, H.T., 1987. Microbial Reduction of StructuralIron(III) in Smectites. Soil Science Society of America Journal 51, 1663–1665.

Summons, R.E., Amend, J.P., Bish, D., Buick, R., Cody, G.D., Des Marais, D.J., Dromart,G., Eigenbrode, J.L., Knoll, A.H., Sumner, D.Y., 2011. Preservation of Martianorganic and environmental records: final report of the Mars BiosignatureWorking Group. Astrobiology 11, 157–184.

Suquet, H., de la Calle, C., Pezerat, H., 1975. Swelling structural organization ofsaponite. Clays and Clay Minerals 23, 1–9.

Theng, B.K.G., 1974. The Chemistry of Clay-Organic Reactions. Halsted Press.Tosca, N.J., Knoll, A.H., 2009. Juvenile chemical sediments and the long term

persistence of water at the surface of Mars. Earth and Planetary Science Letters286, 379–386.

Tosca, N.J., Milliken, R.E., Michel, F.M., 2008. Smectite Formation on Early Mars:Experimental Constraints. Workshop on Martian Phyllosilicates: Recorders ofAqueous Processes, Abstract #7030.

Vázquez-Mayagoitia, A., Horton, S.R., Sumpter, B.G., Sponor, J., Sponer, J.E., Fuentes-Cabrera, M., 2011. On the stabilization of ribose by silicate minerals. Astro-biology 11, 115–121.

Velde, B., 1995. Origin and Mineralogy of Clays. Springer-Verlag.Velde B., 2003. Fast action in soils: Kinetics of clay mineral change under conditions

of agriculture and the role of organic matter. In: Proceedings of the 10thConference of the European Clay Groups Association, 278.

Velde, B., Meunier, A., 2008. The origin of clay minerals in soils and weatheredrocks. Springer-Verlag.

Vorhies, J.S., Gaines, R.R., 2009. Microbial dissolution of clay minerals as a source ofiron and silica in marine sediments. Nature Geoscience 2, 221–225.

Wattel-Koekkoek, E.J.W., Buurman, P., Van Der Pilcht, J., Van Breemen, N., 2003.Mean residence time of soil organic matter associated with kaolinite andsmectite. European Journal of Soil Science 54, 269–278.

Whitney, G., 1990. Role of water in the smectite-to-illite reaction. Clays and ClayMinerals 38, 343–350.

Wilkins, R.W.T., Ito, J., 1976. Infrared spectra of some synthetic talcs. AmericanMineralogist 52, 1649–1661.

Wray, J.J., Ehlmann, B.L., Squyres, S.W., Mustard, J.F., Kirk, R.L., 2008. Compositionalstratigraphy of clay-bearing layered deposits at Mawrth Vallis, Mars. Geophy-sical Research Letters, 35, http://dx.doi.org/10.1029/2008GL034385.

Wray, J.J., Murchie, S.L., Squyres, S.W., Seelos, F.P., Tornabene, L.L., 2009. Diverseaqueous environments on ancient Mars revealed in the southern highlands.Geology 39, 1043–1046.

Zhang, G., Dong, H., Kim, J.W., Eberl, D.D., 2007. Microbial reduction of structuralFe3+ in nontronite by a thermophilic bacterium and its role in promoting thesmectite to illite reaction. American Mineralogist 92, 1411–1419.

http://dx.doi.org/10.1029/2009JE003351

http://dx.doi.org/10.1029/2009JE003351

http://dx.doi.org/10.1029/2009JE003351

http://dx.doi.org/10.1029/2009JE003351



















http://dx.doi.org/10.1029/2006JE002831

http://dx.doi.org/10.1029/2006JE002831

http://dx.doi.org/10.1029/2006JE002831

http://dx.doi.org/10.1029/2006JE002831


















http://dx.doi.org/10.1051/0004-6361:200810150

http://dx.doi.org/10.1051/0004-6361:200810150

http://dx.doi.org/10.1051/0004-6361:200810150







































































http://dx.doi.org/10.1029/2008GL034385

http://dx.doi.org/10.1029/2008GL034385

http://dx.doi.org/10.1029/2008GL034385







Documents

Planetary and Space Science · 18 m/pixel in the full resolution targeted (FRT) mode and at ∼40 m/pixel in the half-resolution short or long (HRS/HRL) mode (Murchie et al., 2009c)