Constraints on the composition and petrogenesis of the ...these compositions might have arisen on...

Constraints on the composition and petrogenesis of the Martian crust

Harry Y. McSween Jr.Department of Earth and Planetary Sciences, University of Tennessee, Knoxville, Tennessee, USA

Timothy L. GroveDepartment of Earth, Atmospheric and Planetary Sciences, Massachusetts Institute of Technology, Cambridge,Massachusetts, USA

Michael B. WyattDepartment of Geological Sciences, Arizona State University, Tempe, Arizona, USA

Received 27 August 2003; revised 10 October 2003; accepted 29 October 2003; published 13 December 2003.

[1] Spectral interpretation that silicic rocks are widespread on Mars implies that Earth’sdifferentiated crust is not unique. Evaluation of observations bearing on the compositionof the Martian crust (Martian meteorite petrology and a possible crustal assimilant,analysis of Mars Pathfinder rocks, composition of Martian fines, interpretation ofspacecraft thermal emission spectra, and inferred crustal densities) indicates that the crustcan be either basalt plus andesite or basalt plus weathering products. New calculatedchemical compositions for Thermal Emission Spectrometer (TES) global surface unitsindicate that surface type 1 has basaltic andesite composition and surface type 2 has thecomposition of andesite. If these materials represent volcanic rocks, their calc-alkalinecompositions on a FeO*/MgO versus silica diagram suggest formation by hydrous meltingand fractional crystallization. On Earth, this petrogenesis requires subduction, and it maysuggest an early period of plate tectonics on Mars. However, anorogenic productionof andesite might have been possible if the primitive Martian mantle was wet.Alternatively, chemical weathering diagrams suggest that surface type 2 materials couldhave formed by partial weathering of surface type 1 rocks, leading to depletion insoluble cations and mobility of silica. A weathered crust model is consistent with theoccurrence of surface type 2 materials as sediments in a depocenter and with the alphaproton X-ray spectrometer (APXS) analysis of excess oxygen suggesting weathering rindson Pathfinder rocks. If surface type 1 materials are also weathered or mixed withweathered materials, this might eliminate the need for hydrous melting, consistent with arelatively dry Martian mantle without tectonics. INDEX TERMS: 6225 Planetology: Solar

System Objects: Mars; 5480 Planetology: Solid Surface Planets: Volcanism (8450); 3672 Mineralogy and

Petrology: Planetary mineralogy and petrology (5410); 1020 Geochemistry: Composition of the crust; 1060

Geochemistry: Planetary geochemistry (5405, 5410, 5704, 5709, 6005, 6008); KEYWORDS:Mars, crust, basalt,

andesite, weathering, thermal emission spectroscopy

Citation: McSween, H. Y., Jr., T. L. Grove, and M. B. Wyatt, Constraints on the composition and petrogenesis of the Martian crust,

J. Geophys. Res., 108(E12), 5135, doi:10.1029/2003JE002175, 2003.

1. Introduction

[2] A comparison of surface materials on the Earth andMoon illustrates that planetary differentiation can followvarying paths that lead to quite different crustal composi-tions. Until recently, the Earth’s silicic continental crust(mostly a result of hydrous melting and fractionation insubduction zones) was thought to be geochemically unique[Rudnick, 1995]. However, the analysis of rocks havingchemical compositions similar to andesite at the Mars Path-finder landing site [Rieder et al., 1997;McSween et al., 1999;Waenke et al., 2001; Foley et al., 2003] suggests that the

Martian crust contains silicic rocks, and the Mars GlobalSurveyor Thermal Emission Spectrometer (MGS-TES)mapped abundance of a global spectral unit interpreted tobe andesitic [Bandfield et al., 2000; Hamilton et al., 2001]may imply that such silicic rocks are widespread. Thishypothesis challenges our understanding of petrogenesis ona world seemingly without plate tectonics.[3] Mars has a voluminous crust, averaging �50 km

thickness [Zuber, 2001] and comprising >4% of the plane-tary volume (compared to�1% crust on the Earth). A crustaldichotomy separates thick, ancient (Noachian) highlandscrust in the southern hemisphere from thinner, less denselycratered lowlands in the north. The lowlands are coveredwith younger (Hesperian) sediments of the Vastitas Borealis

JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 108, NO. E12, 5135, doi:10.1029/2003JE002175, 2003

Copyright 2003 by the American Geophysical Union.0148-0227/03/2003JE002175$09.00

9 - 1

Formation [Scott et al., 1987; Tanaka et al., 1988, 2003], butthe density of large, incompletely buried craters is compa-rable to that in the highlands [Frey et al., 2002], suggestingthat the age of the northern lowlands basement is alsoNoachian. The Tharsis rise, a huge dome containing massiveshield volcanoes, separates these terranes along part of theirjoin. The volcanic surfaces on and around these volcanoesare relatively young (Amazonian), although Tharsis itselfhas been a locus of plume volcanism for billions of years[Phillips et al., 2001]. The lithosphere under this bulge maybe >100 km thick if the load is supported isostatically[Solomon and Head, 1982]. Elysium is another, smallerplume containing large volcanoes and Amazonian flows.Figure 1, which schematically illustrates salient features ofthe various major subdivisions of the Martian crust, willserve as a useful guide as we consider constraints on crustalcomposition and origin.[4] Early formation of the bulk of the Martian crust is

inferred from its bombardment history [Hartmann andNeukum, 2001] and the �4.5 Ga measured crystallizationage of ALH84001 [Nyquist et al., 2001], the only Martianmeteorite that directly sampled the ancient crust. The�4.0 Ga 40Ar/39Ar age of this meteorite is thought to reflectthe late heavy bombardment [Ash et al., 1996]. The extrac-tion of incompatible elements during formation of this earlycrust depleted mantle source regions that later melted toproduce younger Martian meteorites, which still carry thegeochemical signature of this 4.5 Ga fractionation event intheir strontium [Borg et al., 1997] and lead [Chen andWasserburg, 1986] radiogenic isotope systems. Moreover,the former existence of short-lived radionuclides like 146Smin the Mars mantle, as documented in Martian meteorites[Harper et al., 1995; Borg et al., 1997], demands earlycrustal differentiation. Some additional crust formationoccurred during the Hesperian, when lavas flooded thenorthern plains to depths of 1–2 km [Frey et al., 2002],and significant amounts of accompanying plutonic rockmust also have been added to the northern plains crustduring this time.[5] Mars exhibits a rich variety of volcanic landforms,

leading to speculation that its crust consists mostly ofigneous rocks [e.g., Greeley and Spudis, 1981]. If layeringin the walls of Valles Marineris consists entirely of lavaflows [McEwen et al., 1999] and such layering wereglobally distributed, it would suggest that virtually the entirecrustal thickness is of igneous origin. However, Malin andEdgett [2000, 2001] have interpreted some Valles Marinerislayers as well as thick (�10 km) layers covering other partsof the Martian surface as sedimentary rocks, albeit likelyderived from igneous precursors. There is little doubt thatboth volcanic and sedimentary units and landforms arecommon on Mars, but the relative proportions of thesematerials within the crust remains controversial.[6] A related controversy is whether the Martian crust

experienced chemical (as opposed to only mechanical)weathering processes with accompanying chemical fractio-nations. The preservation of igneous mineralogy on Marsis indicated by the spectral identification of pyroxenes,plagioclase, and sometimes olivine in regions not blanketedby dust [Mustard et al., 1997; Bandfield et al., 2000;Hamilton et al., 2001; Bandfield, 2002]. However, a varietyof alteration phases (e.g., sheet silicates, amorphous silica,

zeolites, and palagonites) in modest proportions have beensuggested as analog components in thermal emission spec-tra of low-albedo Martian surface materials [Wyatt andMcSween, 2002; Kraft et al., 2003; Ruff and Christensen,2003; Morris et al., 2003]. Evidence for crystalline phyllo-silicates has not been observed in visible/near-infrared(VISNIR) spectra, leading to speculation that any hydroussilicates must be poorly crystalline or amorphous [e.g.,Bell et al., 2000]. Hydrogen in equatorial regions analyzedby the Mars Odyssey Gamma-Ray Spectrometer [Boyntonet al., 2002] can be interpreted as mineral-bound OH,supporting the idea of chemical weathering. Pervasiveerosional striping and widespread overland deposition, asinferred from MGS MOC images [Malin and Edgett, 2001],would have facilitated both chemical and mechanical weath-ering of volcanic rocks.[7] Despite decades of spectral mapping of the planet’s

surface by orbiting spacecraft, chemical analysis of surfacematerials by several landers and rovers, and laboratory studyof several dozen Martian meteorites, the bulk composition ofthe Martian crust remains undefined and its origin poorlyunderstood. Here, we critically evaluate constraints on thechemistry and petrology of the ancient crust, derive newgeochemical data from TES spectroscopy for different partsof the crust, and explore how crustal components havingthese compositions might have arisen on Mars.

2. Constraints on Mars Crust Composition

2.1. Petrology of Martian Meteorites

[8] ALH84001 offers an extremely limited and biasedsampling of the ancient crust. This �4.5 Ga-old ultramafic(orthopyroxenite) cumulate cannot represent the bulk com-position of the highlands, because partial melting of anultramafic mantle cannot yield an ultramafic crust. Instead,it has been suggested that ALH84001 formed by fractionalcrystallization of a basaltic parent magma [Mittlefehldt,1994]. The apparent failure of impacts to dislodge moremeteorites from the ancient crust may stem from an inabilityof old crust to transmit the requisite shock waves because ofscarcity of coherent rocks within these terranes [Head et al.,2002a; McSween, 2002].[9] The other Martian meteorites, collectively called

SNCs (an acronym for shergottite, nakhlite, chassignite),have young crystallization ages, ranging from 175 to1300 Ma [Nyquist et al., 2001]. Shergottites are subdividedinto basaltic shergottites (tholeiitic basalts, sometimeswith modest amounts of cumulus pyroxenes), olivine-phyricshergottites (basalts containing olivine xenocrysts or phe-nocrysts), and lherzolitic shergottites (plagioclase-bearingperidotites). Nakhlites (olivine-bearing clinopyroxenites)are similar to pyroxenites in some terrestrial komatiiteflows, and Chassigny (a dunite) is chronologically andgeochemically linked to the nakhlites. All these meteoritesare basaltic rocks or plutonic cumulates likely formed byfractional crystallization of basaltic magmas. Compositionsof basaltic shergottites (including olivine-phyric samples)and nakhlites plot within the field of basalt on a chemicalclassification diagram for volcanic rocks (Figure 2).[10] The young crystallization ages of SNCs point to their

derivation from young volcanic centers, probably Tharsis orElysium, apparently the only regions on Mars geologically

9 - 2 MCSWEEN ET AL.: MARTIAN CRUST COMPOSITION AND PETROGENESIS

recent enough to host these rocks. Attempts to locate specificlaunch sites for Martian meteorites using TES spectra havenot been successful [Hamilton et al., 2003], because Tharsis,Elysium, and perhaps other young volcanic terranes aremantled by thick blankets of dust. Earlier studies concludedthat visible/near-infrared (VISNIR) spectra of dark regions(without dust cover) on Mars were similar to basalticshergottites [Mustard et al., 1997]. However, VISNIR spec-tra are primarily sensitive to the presence of ferromagnesian

minerals, and the similarity in this case refers to pyroxenecompositions rather than to the bulk mineralogy.[11] The apparent absence of rocks other than basalts or

their plutonic derivatives among young Martian meteoritessuggests that the crust comprising the young volcaniccenters is basaltic. However, the fact that many Martianmeteorites are partial cumulates demonstrates that fraction-ation did occur, possibly yielding more silicic residualmagmas. Attempts to constrain lava rheologies, and therebyinfer lava compositions, from the eruption styles of Tharsisand Elysium volcanoes allow a wider range of magmacompositions [e.g., Zimbelman, 1985; Catermole, 1987;Baloga et al., 2003], but these models suffer from ambigu-ities related to magma effusion rate, crystallinity, and otherfactors that influence flow rheology.[12] Although SNC meteorites have provided critical

insights into the timing of crust formation and the geo-chemical nature of the mantle source region after early crustextraction, they are not samples of the ancient crust and thuscannot provide direct information on its composition. Laterin this paper, we will introduce evidence that the ancientcrust is distinctly different from Martian plume magmatismas revealed by SNCs.

2.2. Crustal (?) Reservoir Sampled byShergottite Magmas

[13] The parental magmas for basaltic and olivine-phyricshergottites sampled two distinct geochemical reservoirs.Correlations between initial radiogenic isotope ratios(87Sr/86Sr, 143Nd/144Nd, 176Hf/177Hf) and fractionations ofrare earth elements (REE) could indicate incorporation

Figure 1. Schematic illustration of major subdivisions ofthe Martian crust. The exposed southern highlands and thebasement beneath the Hesperian cover in the northernlowlands are of Noachian age. Volcanism associated withthe Tharsis plume produced a very thick crust and extendedinto the Amazonian period. Crustal thickness estimates arefrom Solomon and Head [1982] and Zuber [2001].

Figure 2. Total alkalis versus silica classification diagram for volcanic rocks [Le Bas et al., 1986],showing the compositions of basaltic shergottites (including olivine-phyric shergottites) and nakhlites[Lodders, 1998; Dreibus et al., 2000; Folco et al., 2000; Rubin et al., 2000; Barrat et al., 2002; Taylor etal., 2002; Imae et al., 2003], two calibrations of the Mars Pathfinder dust-free rock (point with loweralkalis from Waenke et al. [2001], point with higher alkalis from Foley et al. [2003]), and MGS-TES-derived chemical compositions (Table 4) for surface types 1 and 2 of Bandfield et al. [2000] (B),Hamilton et al. [2001] (H), Wyatt and McSween [2002] (W), and our new estimates based on an extendedspectral range (X-boxes). The same symbols are used in subsequent diagrams.

MCSWEEN ET AL.: MARTIAN CRUST COMPOSITION AND PETROGENESIS 9 - 3

of varying amounts of ancient, highly radiogenic, light REE-enriched crust by melts from a depleted mantle [Borg et al.,1997;McSween, 2002]. The assimilated material was similarin many respects to the lunar KREEP component. Alterna-tively, these correlations could indicate mixing of materialsfrom complementary enriched and depleted reservoirs withinthe Martian mantle [Wadhwa and Grove, 2002; Borg et al.,2003]. In this case, the depleted and enriched regions musthave been isolated by an early differentiation event, possiblycreating a shallow, enriched mantle and a deep, depletedmantle. A correlation between the degree of geochemicalcontamination and magmatic oxidation state [Wadhwa,2001; Herd et al., 2002] is consistent with crustal assimila-tion, although recent evidence shows that the depletedsource region for nakhlites could also be derived from anoxidized mantle reservoir [Wahdwa and Grove, 2002].[14] For the moment, let us assume that the assimilant

was crust. Can we use the composition of this component toconstrain the nature of the ancient crust through which theshergottites erupted? Unfortunately, it is easier to determinethe isotopic and trace element compositions of this compo-nent than its major element abundances or petrologicidentity. It is not even clear if this component was actuallyrock or a fluid that either metasomatized the shallow mantleor scavenged solutes from the crust. If rock was assimilated,the ancient crust probably has a basaltic composition;assimilation of andesite would have increased the silicacontents of the resulting contaminated magmas. Figure 3ashows silica abundances in shergottite melts plotted versusLa/Yb (a proxy for the abundance of the assumed crustalcomponent). Although most shergottites show a positivecorrelation, as would be expected if the assimilant weresilicic rock, the Los Angeles shergottite is inconsistent withthis model. If the ancient crust can be represented by theMars Pathfinder dust-free rock (see below), incorporation ofthis component should also have resulted in lower MgO/Al2O3 (Figure 3b), which is not observed. (Note: La/Yb forthe Pathfinder rock is unknown but is assumed in Figure 3to be high, as appropriate for fractionated crust.)

2.3. In Situ Chemical Analysis ofMartian Crustal Rocks

[15] The Mars Pathfinder rover analyzed five rocks usingan alpha proton X-ray spectrometer (APXS). PreliminaryX-ray mode analyses of rocks at the Pathfinder site [Riederet al., 1997] have now been revised [Waenke et al., 2001;Foley et al., 2003], owing to the inclusion of alpha modedata and to differences in conditions under which labora-tory calibrations and Mars measurements were made.Element concentrations plotted versus sulfur in rocks yieldstraight lines, with soils clustering at the sulfur-rich ends ofthe rock arrays. These trends are interpreted as mixing linesbetween the compositions of rock and dust. Imageryindicates that rocks at the Pathfinder site are partly coatedwith red dust, and a correlation between sulfur content andthe red/blue (750/440 nm) reflectance spectra ratio ofAPXS-analyzed rocks [McSween et al., 1999] reinforcesthe conclusion that rock analyses are contaminated bysulfur-rich dust. Extrapolation of the mixing lines to lowsulfur (Waenke et al. [2001] used 0.3 wt.% S, based onbasaltic shergottite sulfur contents) yields the dust-free rockcomposition.

[16] The SiO2 concentration (57 ± 6 wt.% fromWaenke etal. [2001]; 57.7 ± 1.5 wt.% from Foley et al. [2003]) for thePathfinder dust-free rock plots on the low-silica boundary ofthe andesite field in a classification diagram (Figure 2). Thepreferred interpretations of this rock composition are thatit is volcanic, or is a clastic sedimentary rock composedof volcanic fragments [McSween et al., 1999; Waenke etal., 2001]. However, McSween et al. [1999] also consid-ered the possibility that the dust-free rock composition

Figure 3. Major element compositions for basaltic/oli-vine-phyric shergottites appear to be decoupled from traceelement (La/Yb) data, the latter interpreted to reflect varyingdegrees of assimilation of light REE-enriched crust. Silicacontents for NWA480 and NWA856 are calculated bydifference from the sum of other analyzed oxides.Assimilation of the Mars Pathfinder dust-free rock compo-sition (57% SiO2, 0.14 MgO/Al2O3, presumed to have highLa/Yb) would elevate SiO2 and lower MgO/Al2O3. Datasources as in Figure 2.


might represent a weathered rock surface. Under terrestrialweathering conditions silica is usually leached from basalt,but some basalts exposed to semi-arid conditions develophydrous, silica-rich coatings [Farr and Adams, 1984;Crisp et al., 1990]. Altered silicic coatings on Pathfinderrocks would be consistent with their photometric proper-ties, which may imply varnished rinds [Johnson et al.,1999]. Bishop et al. [2002] developed a model for theformation of Martian rock varnish, involving chemicalreactions between rocks and the dust that settles on them.Foley et al. [2003] noted that plots of alkalis versus sulfurin Pathfinder rocks show considerable scatter, perhapssuggesting mobilization of soluble elements during weath-ering. APXS alpha mode analyses of oxygen [Foley et al.,2003] suggest that the Pathfinder dust-free rock containsmore oxygen than can be accounted for by stoichiometriccombination with its cations. The excess oxygen inShark, the rock with least dust cover, is equivalent to3.3 ± 1.3 wt.% H2O, assuming that half the iron is ferric.This water abundance is very high for igneous rocks, and

may be more consistent with the hypothesis that Pathfinderrocks have weathered, hydrous coatings.

2.4. Martian Surface Fines as Proxies forCrust Composition

[17] The pervasive dust that covers the Martian surface isthought to have been globally homogenized by winds [e.g.,McCord et al., 1982], which may account for the compo-sitional similarity of soil deposits at landing sites separatedby thousands of kilometers [Clark et al., 1982; Waenke etal., 2001]. Fine-grained sediments are commonly used toestimate the composition of the Earth’s crust [McLennanand Taylor, 1984], and surface fines might likewise providea critical constraint on compositions of the dominant crustalrocks on Mars.[18] Originally, the compositional similarity between

Viking soils and basaltic shergottites was cited as evidencethat the soils formed from basalts [Toulmin et al., 1977].Following the discovery of rocks having andesitic compo-sitions by Mars Pathfinder, a number of workers reinter-preted Pathfinder and Viking soil compositions as mixturesof basaltic (SNC) and andesitic materials in roughly equalproportions [Larsen et al., 2000; Morris et al., 2000;Waenke et al., 2001]. At face value, this would seem tobe evidence for the existence of significant amounts ofsilicic crust. Such a model presumes that soils formedby mechanical weathering of rocks, without significantchemical modification.[19] Alternatively, analyzed Martian soils could represent

mixtures of a common (globally homogenized) dust com-ponent with varying amounts of local (andesitic) rockparticles at the Pathfinder site and with sulfate cements atthe Viking sites [McSween and Keil, 2000]. Because thePathfinder APXS analyses of soils have recently beenrecalibrated [Waenke et al., 2001; Foley et al., 2003], wehave replotted these compositions to see if the trends notedby McSween and Keil [2000] persist (Figure 4). The recali-brated Pathfinder soil analyses are not as readily interpretedas dust with admixed local rock (Figure 4), but it is stillpossible to estimate an average soil composition fromuncemented Pathfinder and Viking soils. This composition,which we assume represents the global dust, is presented inTable 1 (the calculation procedure and assumptions are alsodescribed in Table 1) and shown by Xs in Figure 4. Thecommon dust composition itself can be modeled as basaltthat has undergone a moderate degree of chemical weather-ing [McSween and Keil, 2000]. Various chemical weatheringmechanisms (palagonitization, hydrothermal alteration,reactions of rocks with acid fog formed by volcanic exha-lations) have been suggested as possible origins for Martiansoils (summarized by Bell et al. [2000]). However, thehypothesis that the global dust may have formed by simplemixing of basaltic and andesitic components does not seemto be tenable, given the mismatch for silicon, iron, andpotassium in mixing calculations (Figure 5).

2.5. Thermal Emission Spectroscopy of the Crust

2.5.1. Surface Compositions[20] Two distinct global surface spectral signatures have

been identified in low-albedo regions on the Martian surface[Christensen et al., 2000a; Bandfield et al., 2000] usingatmospherically corrected thermal emissivity data [Smith et

Figure 4. Variation diagrams illustrating the global dustcomposition (X), as well as the effects of admixed andesiticrock chips in Pathfinder soils and sulfate salt cement inViking soils. The calculated dust composition (Table 1) mayrepresent a nearly common chemical component at thePathfinder and Viking 1 sites. Representative error bars areshown for several soil analyses from the Viking 1 andPathfinder sites.


al., 2000] from TES. The surface type 1 spectral end-member has been interpreted as unaltered basalt character-ized by high deconvolved abundances of plagioclase andclinopyroxene [Christensen et al., 2000a; Bandfield etal., 2000; Hamilton et al., 2001]. The surface type 2spectral end-member has been variously interpreted asunaltered basaltic andesite or andesite [Bandfield et al.,2000; Hamilton et al., 2001] or as partly altered basalt[Wyatt and McSween, 2002; Morris et al., 2003]. Theandesitic composition is characterized by high deconvolvedabundances of plagioclase and high-silica volcanic glass[Bandfield et al., 2000; Hamilton et al., 2001]. The partlyaltered basalt is characterized by high modal abundances ofplagioclase and a variety of alteration phases (sheet silicates,silica coatings, and palagonite) and low modal pyroxene[Wyatt and McSween, 2002; Morris et al., 2003]. Detectableabundances of hematite [e.g., Christensen et al., 2000b],orthopyroxene [e.g., Hamilton et al., 2003; Bandfield,2002], and olivine [e.g., Clark et al., 1982; Hamilton etal., 2003; Bandfield, 2002] have also been identified inregional and local occurrences where surface type 1compositions dominate surface units. The identification ofthese phases may represent unique surface lithologies (i.e.,dunite), or higher abundances of each phase in a basalticsurface unit (i.e., olivine-bearing basalt). Here we focus onsurface type 2, as a huge expanse of andesite (versus alteredbasalt) would significantly influence the bulk compositionof the Martian crust.[21] The initial ambiguity in interpreting the surface type 2

lithology from deconvolved TES mineral abundances arosebecause volcanic siliceous glass (a major component ofandesite) was shown to be spectrally similar to somealteration phases (sheet silicates, amorphous silica coatings,and K-feldspar) over the spectral ranges used in deriving theTES surface spectral end-members [Wyatt and McSween,2002]. Absorption features between 500–550 cm�1 inlaboratory spectra can be used to distinguish well-crystallineclay minerals from high-silica volcanic glass; however, thisregion was excluded by Bandfield et al. [2000] whilederiving the TES spectral end-members because the CO2

atmosphere of Mars is largely opaque near this spectralregion [Wyatt and McSween, 2002]. Significant quantities ofwell-crystalline clay minerals are not indicated by near-infrared observations using Mariner 9 ISM data [Murchie etal., 2000] or telescopic observations [Blaney et al., 2003],and work by Ruff [2003] examining the 500–550 cm�1

region in non-atmospherically corrected TES spectrapoints to the lack of spectral absorption features indicativeof well-crystalline clays. However, these studies do suggestthat poorly crystalline clays and/or other alteration phases

may be permissible, and recent work [Morris et al., 2003;Ruff and Christensen, 2003] has shown palagonites andzeolites to be spectrally similar to high-silica glass. Somehighly shocked feldspars (maskelynites) are also spectrallysimilar to high-silica glass [Johnson et al., 2002]. Theoriginal interpretation of the high-silica glass spectral end-member as a primary volcanic glass [Bandfield et al., 2000]was also shown to be too limited, as deconvolved modalabundances of the natural surface of a terrestrial floodbasalt suggested it could also represent an amorphoushigh-silica alteration product [Wyatt and McSween, 2002].Analyses by Kraft et al. [2003] have further shown that theaddition of high-silica alteration coatings on basalts resultsin a surface that is spectrally similar to andesite.2.5.2. Distributions of Surface Compositions[22] A global view of the distribution of surface type 1

and type 2 materials is shown in Figure 6. The distributionof the surface type 1 (basalt - green) unit is restricted to thesouthern highlands and Syrtis Major regions of Noachian orHesperian age and to a few local occurrences in the northern

Table 1. Recalculated Martian Fines Composition Based on Recalibration of Mars Pathfinder APXS Analyses of Soilsa

Na2O MgO Al2O3 SiO2 P2O5 SO3 Cl K2O CaO TiO2 Cr2O3 MnO Fe2O3

MP(Waenke) 1.1 8.5 7.8 41.5 1.3 6.7 0.5 0.7 6.4 1.0 0.3 0.5 21.8MP(Foley) 2.7 7.3 10.0 40.8 0.8 6.0 0.9 0.5 5.9 0.8 0.3 0.3 21.7MP(fines) 2.1 7.9 8.9 41.1 1.0 6.3 0.7 0.6 6.2 0.9 0.3 0.4 21.8V(fines) 6.0 7.9 46.6 7.2 0.8 6.3 0.7 19.5Global dust 2.1 7.0 8.4 43.9 1.0 6.8 0.7 0.7 6.3 0.8 0.3 0.4 20.7

aMartian fines composition is measured in wt.% oxides. Mars Pathfinder soil averages of Waenke et al. [2001] and Foley et al. [2003], indicated by MP,include all reported soil analyses except A8 (cemented soil). All Pathfinder analyses were normalized to 98% to allow for unanalyzed water (estimated �2wt% [Biemann et al., 1977; Yen et al., 1998]). MP(fines) is an average of the MP(Waenke) and MP(Foley) values. V(fines) is an average of all Viking 1analyses of fines (C-1, C-6, C-7, C-8, C-9) from Clark et al. [1982]. All Viking analyses were normalized to 94.5%, to allow for unanalyzed Na2O, P2O5,Cr2O3, MnO (average Pathfinder values were assumed) and 2% water. Global dust is an average of MP(fines) and V(fines).

Figure 5. Two-component mixing diagram, testing thehypothesis that Martian fines (global dust) formed bycombining basaltic (SNC meteorite) and andesitic (MarsPathfinder dust-free rock) components. The best fit for dust(Table 1, salt-free) has significant discrepancies in silicon,iron, and potassium.


plains [Bandfield et al., 2000; Rogers and Christensen,2003]. The surface type 2 unit (andesite and/or alteredbasalt - red) displays the highest concentrations in theyounger Amazonian-age northern lowlands regions of Acid-alia Planitia and the circumpolar sand seas [Bandfield et al.,2000; Rogers and Christensen, 2003]. These materials inthe northern lowlands have been mapped as the VastitasBorealis Formation [Scott et al., 1987]. Surface type 2compositions are also present in moderate abundances, ormixed with surface type 1, throughout the low-albedosouthern highlands. Blue pixels in Figure 6 representregions covered by a blanket of dust which prohibitsspectral analysis of sand and rock compositions. Thedistribution of the highest concentrations and largest extentsof the two surface spectral units is thus split roughly alongthe planetary topographic dichotomy separating the ancient,heavily cratered crust in the southern hemisphere fromyounger lowland plains in the north.[23] It is important to note that a single interpretation of

surface type 2 spectra may not be warranted everywhere onMars. THEMIS data from Mars Odyssey show adjacentvolcanic units of surface type 1 and 2 materials within theNili Patera caldera [Ruff and Christensen, 2003]. Theseunits have high thermal inertias and probably representoutcrops of lava and/or tuff. For these units an igneousorigin, involving successive eruptions of basaltic and an-desitic magmas, is a reasonable interpretation. Elsewhere,deposits of sediments with surface type 2 spectra may bemore plausibly interpreted as partly weathered basalt. Forexample, Wyatt et al. [2003] described deposits of surfacetype 1 sand dunes on the floors of large craters in OxiaPalus, adjacent to surface type 2 materials on the downwindsides of the crater walls. In this case, surface type 2 can bereadily explained as a finer-grained fraction (containing

some alteration materials) winnowed by winds from thecoarser basaltic sediment on the crater floor.

2.6. Density of the Crust

[24] Using MGS MOLA data, new models of the rela-tionship between gravity and topography [Turcotte et al.,2002] and admittance techniques [McGovern et al., 2002;McKenzie et al., 2002] suggest densities of 2.95–3.15 g/cm3

for parts of the elastic lithosphere of Mars. Turcotte et al.[2002] have argued that the crust is volumetrically equiva-lent to the elastic lithosphere, although other models suggesta thinner crust [Zuber et al., 2001; Nimmo, 2002]. Theinferred density is significantly higher than that estimatedfor the Earth’s continental crust (�2.75 g/cm3), which has anaverage composition of andesite. Although the Martiancrustal density appears to be inconsistent with a dominantlyandesitic crust, it might be consistent with hydrous magma-tism, which could produce dense pyroxenitic cumulates inthe lower crust [Muentener et al., 2000], perhaps resemblingthe ALH84001 orthopyroxenite.[25] Thermal and compositional buoyancy forces in the

mantle source regions of basaltic magmas cause them toascend and erupt on planetary surfaces. On Earth, mid-ocean ridge basalts accumulate as much as several kilo-meters below the level of neutral buoyancy (estimated at100–400 m), suggesting that magma density may notcontrol ascent once the magma reaches the shallow crust[Hooft and Detrick, 1993; Ryan, 1993]. Ultimately,buoyancy is likely to control ascent at deeper levels, butcompositions of the most common magmas from mid-ocean ridges indicate that eruption controls are complex[Grove et al., 1993; Michael and Cornell, 1998] and notsolely a function of density contrast [Stolper and Walker,1980]. Thus it may be unrealistic to use SNC magma

Figure 6. Global simple cylindrical projected TES image of the distribution of surface type 1 (green)and surface type 2 (red) materials on Mars, overlaid on a 128 pixels/deg MOLA DEM. The highestconcentration of surface type 2 materials is in the northern lowlands, generally corresponding to theVastitas Borealis Formation. Blue pixels represent dust-covered regions.


densities to constrain the density (and hence composition)of the Martian crust.

3. Crust Geochemistry From ThermalEmission Spectra

3.1. Background and Method

[26] Volcanic rocks are commonly classified by theirchemical compositions because their modal mineralogiesare not always diagnostic. TES is a mineralogical tool, butit can also provide a means of estimating chemistry.Hamilton and Christensen [2000] demonstrated that thechemical compositions of laboratory-analyzed rocks can beaccurately calculated from deconvolved modal mineralogiesby combining the compositions (wt.% oxides) of thespectral end-members in proportion to their relative mod-eled abundances. Wyatt et al. [2001] further quantifiedthe uncertainties in derived chemical compositions anddemonstrated their use in correctly classifying volcanicrocks. Errors for most oxides, as determined from the Wyatt

et al. [2001] study of terrestrial volcanic rocks, are ± 5%.Derived chemical abundances from thermal emissionspectra are thus a recasting of rock compositions into aform which complements modeled mineral abundances.[27] Hamilton et al. [2001] convolved laboratory spectral

data (2 cm�1 spectral sampling) of rocks from Wyatt et al.[2001] to the lowest spectral resolution of the TES instru-ment (10 cm�1 spectral sampling) and showed that derivedbulk rock chemistries were not significantly degraded.These results demonstrated the feasibility of using similartechniques and classification schemes for TES spectralresolution data. Hamilton et al. [2001] also derived chem-ical compositions of the surface types 1 and 2 globalspectral units and classified them as basalt and andesite,respectively. However, their spectral end-member set wasoptimized for igneous rocks.[28] Here, we estimate and compare surface type 1 and 2

chemical compositions derived from three previouslypublished modal abundances deconvolved using spectralend-member sets that include a broad range of igneous and

Figure 7. Comparison of TES spectra for (a) surface type 1 and (b) surface type 2 with modeled spectralfits (offset by 0.028 emissivity for clarity) and mineral abundances produced by linear deconvolutionfrom Bandfield et al. [2000], Hamilton et al. [2001],Wyatt and McSween [2002], and this study (Table 3).Mineral abundances for surface type 1 are consistent with basaltic rocks, whereas surface type 2abundances are consistent with either andesite or partly weathered basalt.


sedimentary minerals [Bandfield et al., 2000; Hamilton etal., 2001; Wyatt and McSween, 2002]. Spectral fittingfor those studies was constrained to 1280–400 cm�1,although TES data actually cover the wave number rangeof 1650–233 cm�1. The atmospheric correction used toderive the TES surface spectra [Bandfield et al., 2000] didnot include the high wave number range of TES datadue to numerous water vapor and minor CO2 features.Furthermore, there are no fundamental silicate features inthe 1650–1400 cm�1 region. The wave number rangeof 400–233 cm�1 was not used for deconvolutions inprevious studies because mineral end-member spectrain the ASU Thermal Emission Spectroscopy library[Christensen et al., 2000c] only extended to 400 cm�1.This laboratory now has a spectrometer that covers thefull TES wavelength range (1650–200 cm�1) and emis-sivity spectra have been measured for all end-members,enabling us to expand the spectra range used for decon-volution of TES surface type 1 and 2 materials. Thus, inthis study, we also estimate surface type 1 and 2 compo-sitions from new linear deconvolutions that cover theexpanded spectral range of 400–233 cm�1. Modeledspectral fits, deconvolution modal abundances, andderived chemistries from each of the previously published

studies and our new work are examined to constrainMartian surface compositions.

3.2. Model Results and Classification

[29] Figures 7a and 7b compare spectral fits for lineardeconvolutions of surface types 1 and 2 spectra, respectively.Bandfield et al. [2000] used 45 spectral end-members repre-senting igneous, sedimentary, and metamorphic minerals,whereas Hamilton et al. [2001] used a narrower range of29 mineral spectra common in unweathered basalts andandesites. Wyatt and McSween [2002] used 39 spectral end-members representing igneous and alteration minerals inpartly weathered basalts. In this study, we use 46 spectralend-members representing a similar wide range of igneousand sedimentary minerals in unweathered and weatheredbasalts and andesites (Table 2). Overall, spectral fits producedby the linear deconvolution algorithm using the differentend-member sets are very good (Figures 7a and 7b),suggesting major rock phases are well represented in theend-member libraries and that they provide acceptable fitsto the rock types in this study. Spectral fits of the surfacetype 1 spectrum show low RMS values of 0.0018 [Bandfieldet al., 2000], 0.0026 [Hamilton et al., 2001], 0.0018 [Wyattand McSween, 2002], and 0.0022 (this study). Spectral fits of

Figure 7. (continued)


the surface type 2 spectrum show low RMS values of0.0009 [Bandfield et al., 2000], 0.0023 [Hamilton et al.,2001], 0.0014 [Wyatt and McSween, 2002], and 0.0019 (thisstudy). Figures 7a and 7b also demonstrate that the extendedspectral range used in this study is well modeled anddoes not adversely affect the overall quality of modeledspectra. Deconvolved mineral abundances from each of theprevious studies and new mineral abundances from thisstudy are listed in Table 3 and shown in Figures 7a and 7b.The detection limit for mineral abundances deconvolvedfrom TES spectra is 10–15 vol.% based on instrumentuncertainties [Christensen et al., 2000a], errors associatedwith atmospheric corrections [Bandfield et al., 2000; Smithet al., 2000], and limits of the deconvolution technique[Ramsey and Christensen, 1998]. Surface type 1 decon-volved mineral abundances for all end-member sets aresimilar to within the 10–15 vol.% TES uncertainty and the5–10 vol.% absolute uncertainty that has been associatedwith all modeled mineral abundances based on compar-isons with petrographic point-counting modes [Feely andChristensen, 1999; Hamilton and Christensen, 2000] andhigh-resolution electron microbe phase mapping techni-ques [Wyatt et al., 2001]. Surface type 2 deconvolved

mineral abundances for all end-member sets are alsosimilar to within the 10–15 vol.% TES uncertainty, exceptfor feldspar abundances from Hamilton et al. [2001] andthis study which represent the relative maximum andminimum modeled feldspar abundances. The low modeledfeldspar abundance for surface type 2 in this study, andincrease in the total of minor phases modeled well belowTES detectability limits, may result from the extendedwave number range used in deconvolution. The end-member sets that include a variety of alteration phases(phyllosilicates, carbonates, silica) have lower RMS errorsthan the Hamilton et al. [2001] fits, which focused almostentirely on igneous phases. These results suggest thatsmall to modest amounts of alteration minerals may bepresent in both surface types. Carbonate abundance ismodeled well below TES detectability limits, in agreementwith the conclusion of Bandfield [2002] that carbonatesare not detectable in low-albedo regions.[30] Chemical compositions calculated from modeled

mineral abundances for each of the spectral end-membersets are presented in Table 4. The derived chemicalcompositions plotted in Figures 2, 8, and 9 are calculatedon a H2O-free and CO2-free basis (also presented in

Table 2. Spectral End-Members Used for New Deconvolutions of MGS-TES Emissivity Dataa

Feldspars Pyroxenes Phyllosilicates Olivines

Albite WAR5851 Enstatite HS9.4B Muscovite WAR5474 Forsterite AZ01Oligoclase WAR5804 Bronzite NMNH93527 Biotite BUR840 Fayalite WARRGFAY01Andesine BUR240 Diopside WAR6454 Phlogopite HS23.3BLabradorite WARRGAND01 Hedenbergite DSMHED01 Serpentine HS8.4BLabradorite WAR4524 Augite BUR620 Serpentine BUR1690Bytownite WAR1384 Augite NMNH9780 Antigorite NMNH47108Anorthite WAR5759 Pigeoniteb Ca-montmorillonite STX1Anorthite BUR340 Nontronite WAR5108Microcline BUR3460 Saponite ASUSAP01Microcline BUR3460A Illite IMt2Anorthoclase WAR0579 Fe-smectite SWA1

Chlorite WAR1924Na-montmorillonite SWY2

Amphiboles Glasses Carbonates Sulfates

Actinolite HAS116.4B Si-K glassb Calcite MLC10 Anhydrite MLS9Mg-hornblende WAR0354 Basaltic glassb Dolomite C28 Gypsum MLS6

Silica glassb Magnesite C60Siderite C62Aragonite C11

aNumbers indicate ASU spectral library phase.bSpectra of pigeonite and glasses are from Wyatt et al. [2001].

Table 3. Modeled Phase Abundances for MGS-TES Surface Types 1 and 2 Materialsa

Feldspars Pyroxenes Glass Sheet Silicates Other

Surface Type 1Bandfield et al. [2000] 49 29 0 17 6Hamilton et al. [2001] 55 29 9 5 2Wyatt and McSween [2002] 33 41 0 14 12This study 35 29 2 22 14

Surface Type 2Bandfield et al. [2000] 33 10 23 17 17Hamilton et al. [2001] 49 8 28 8 7Wyatt and McSween [2002] 39 16 0 31 14This study 18 8 34 18 24

aAll mineral groups have detection limits of �10–15 vol.%. In modeled results, feldspars are dominated by plagioclase,pyroxenes by high-Ca pyroxene, glass by Si-K glass, and sheet silicates by smectite. Other category includes the sums ofcarbonates and sulfates individually modeled well below detection limits.


Table 4). These figures explore the likelihood that surfacetypes 1 and 2 represent volcanic rock compositions. Thenormalization has the effect of increasing the other oxidecomponents, especially silica contents. Surface type 2 decon-volutions typically contain more hydrous minerals andthus their derived compositions contain more water; conse-quently, their silica contents are increased more than forsurface type 1. Figures 10 and 11, which explore the possi-bility that the derived compositions represent partly weath-ered volcanic rocks, are unaffected by this normalization.[31] Surprisingly, there is relatively little difference

between chemical compositions estimated using the variousspectral end-member sets. Using the total alkalis versus silicaclassification of Le Bas et al. [1986], all surface type 1compositions plot within the field of basaltic andesite andare clearly distinct from Martian meteorites (Figure 2).Surface type 2 compositions plot within the andesite field(Figure 2) and near the Mars Pathfinder dust-free rockcomposition (in this figure, but not necessarily in otherdiagrams). However, surface type 2 compositions that arenot recalculated as H2O-free would plot within the basalticandesite field, closer to surface type 1 compositions. Theclustering of compositions suggests that chemistry derivedfrom TES data is relatively robust, although we caution thatthermal emission spectra sample only the outer few hundredmicrons of grains.

4. Petrogenesis of Crustal Materials

[32] Martian surface compositions could reflect eitherigneous materials or a combination of basalt and alterationproducts. Below, we consider each possibility.

4.1. Origin of Andesitic Magmas

[33] Miyashiro [1974] introduced the FeO*/MgO versusSiO2 diagram (Figure 8) (* indicates that all iron is reportedas FeO) to delineate the tholeiitic (TH) and calc-alkaline

(CA) magmatic trends. In this diagram, anhydrous fractionalcrystallization paths (tholeiitic trends) are nearly vertical, asillustrated by arrows in the TH field showing the trends ofGalapagos mid-ocean ridge lavas (kinked arrow, analyzedby Juster et al. [1989]) and the dry liquid line of descent forbasaltic shergottites (short arrow, calculated using theMELTS program by Hale et al. [1999]). In contrast,fractional crystallization under hydrous conditions followsdistinctive paths (calc-alkaline trends) that lead to highersilica contents, as illustrated by the diagonal arrows in theCA field [Sisson and Grove, 1993; Grove et al., 2003]. Theeffect of H2O is to expand the primary phase volumesfor olivine and clinopyroxene, causing plagioclase to crys-tallize late. Primitive melts produced by low degrees ofhydrous partial melting of mantle peridotite (olivine +orthopyroxene + clinopyroxene + spinel) have compositionsindicated by the large open circle in Figure 8. Thesemagmas would follow a fractionation trend like the diagonalarrow emanating from the circle and paralleling the TH-CAboundary. The heavy gray arrow illustrates the compositionsof liquids derived from higher degrees of hydrous meltingafter clinopyroxene and spinel are exhausted from thesource, leaving only olivine and orthopyroxene (harzburgite)in the mantle residue. The length of the heavy arrowrepresents the 20 to 40% melting interval at 1.2 GPa[Gaetani and Grove, 1998]. This extent of melting isencountered in the mantle wedge in terrestrial subductionzones. Liquids along this arrow are produced by incongruentmelting of orthopyroxene to form olivine + liquid at highwater contents [Grove et al., 2003]. These conditions areonly met on Earth when the mantle wedge above thesubducted slab is fluxed by an H2O-rich fluid derived bydehydration of subducted oceanic plates. Subsequent frac-tional crystallization of these water-rich melts at shallowcrustal levels produces a family of diagonal trends, asillustrated by diagonal arrows representing various terrestrialarc lavas [Grove et al., 2003] in Figure 8.

Table 4. Derived Chemistry for MGS-TES Surface Types 1 and 2 Materialsa

SiO2 TiO2 Al2O3 FeO* MgO CaO Na2O K2O Total

Surface Type 1Uncorrected

Bandfield et al. [2000] 52.2 0.3 17.5 4.9 7.7 10.5 2.3 1.7 97.2Hamilton et al. [2001] 53.5 0.1 15.3 6.8 8.4 10.5 2.9 0.6 98.1Wyatt and McSween [2002] 50.7 0.1 9.7 15.0 9.2 9.7 1.6 0.8 96.8This study 52.3 0.2 13.8 8.7 7.4 10.9 2.0 0.4 95.6

H2O- and CO2-freeBandfield et al. [2000] 53.7 0.3 18.0 5.1 8.0 10.8 2.4 1.8 100.0Hamilton et al. [2001] 54.5 0.1 15.6 6.9 8.5 10.8 2.9 0.6 100.0Wyatt and McSween [2002] 52.5 0.1 10.1 15.5 9.5 10.0 1.6 0.8 100.0This study 54.7 0.2 14.4 9.1 7.8 11.4 2.1 0.4 100.0

Surface Type 2Uncorrected

Bandfield et al. [2000] 56.6 0.1 16.5 5.4 5.4 6.8 2.3 3.0 96.1Hamilton et al. [2001] 57.6 0.1 16.9 3.8 8.6 7.0 2.6 1.3 97.8Wyatt and McSween [2002] 53.8 0.1 16.8 7.0 4.6 9.3 0.9 2.5 95.2This study 55.3 0.7 13.9 10.2 5.4 5.7 1.9 1.4 94.4

H2O- and CO2-freeBandfield et al. [2000] 58.9 0.1 17.2 5.6 5.6 7.1 2.4 3.1 100.0Hamilton et al. [2001] 58.9 0.1 17.3 3.9 8.8 7.1 2.6 1.4 100.0Wyatt and McSween [2002] 56.5 0.1 17.7 7.4 4.8 9.8 1.0 2.6 100.0This study 58.6 0.7 14.7 10.8 5.7 6.0 2.0 1.5 100.0

Errors (1s) 1.4 0.9 1.5 1.2 2.6 0.7 0.4 0.4aMeasured in wt.% oxides.


[34] Our derived compositions for surface type 1 aresimilar to basaltic andesites found in terrestrial subductionzones. On the basis of comparison with terrestrial calc-alkaline lavas (Figure 8), surface type 1 basalts wereproduced by fairly extensive mantle melting under hydrousconditions and at shallow depths (�1 GPa on Earth).Analogous terrestrial melts are in equilibrium with a refrac-tory harzburgite source and contain high magmatic H2Ocontents (4–6 wt.%). Their position above the terrestrialmantle melting arrow could reflect melting of a Martianmantle with higher FeO/MgO. Derived compositions forsurface type 2 are more scattered, but could representmagmas formed by higher degrees of hydrous melting ofthe same mantle (Figure 8). Compositions derived from thisstudy (X-boxes in Figure 8) suggest that andesite havingsurface type 2 composition might have formed by fraction-ation of surface type 1 basaltic andesite magma underhydrous conditions in the shallow Martian crust, althoughthat relationship is not apparent in previously publishedcompositional pairs.[35] Figure 9 illustrates that other chemical trends in

surface type 1 and 2 compositions more closely approxi-mate those of terrestrial calc-alkaline lavas (in this case,lavas from Mt. Shasta [Grove et al., 2002] and hydrouslavas from Medicine Lake volcano [Kinzler et al., 2000])than tholeiitic lavas of the Galapagos spreading center[Juster et al., 1989]. Higher Al2O3 contents are the resultof delayed plagioclase crystallization in hydrous magmas.

Figure 8. FeO*/MgO versus SiO2 diagram showingtholeiitic (TH) and calc-alkaline (CA) fields, shergottiteand nakhlite compositions (data sources as in Figure 2), andTES-derived compositions for surface types 1 and 2. Nearlyvertical black arrows represent tholeiitic (dry) fractionationpaths, and diagonal gray arrows are calc-alkaline (wet)fractionation paths. The large shaded circle at the lower leftcorner represents the composition of a liquid formed byhydrous melting of peridotite (Ol + Opx + Cpx + Sp), andthe heavy gray arrow illustrates how liquids evolve withmore advanced degrees of hydrous melting [after Grove etal., 2003]. Surface type 1 and 2 materials are calc-alkalineand could have formed by hydrous melting and fractionalcrystallization.

Figure 9. Chemical variations (wt.% oxides) in surfacetypes 1 and 2 compared with terrestrial calc-alkaline,tholeiitic, and alkaline lava trends. Calc-alkaline rocks arefrom Mt. Shasta [Grove et al., 2002, 2003] and MedicineLake volcano (hydrous Callahan flows) [Kinzler et al.,2000]; tholeiitic rocks are from the Galapagos spreadingcenter at 85oW [Juster et al., 1989]; and alkaline lavas arefrom the Nandewar Volcano [Abbott, 1969; Stolz, 1985].Representative error bars are shown for TES-derivedcompositions in each panel. Data sources for MarsPathfinder rocks as in Figure 2.


The flat TiO2 pattern in calc-alkaline rocks, reflecting earlycrystallization of Fe-Ti oxide, is also similar.[36] The alkali abundances of surface type 1 and 2 are

clearly subalkaline (Figure 2), although Neksavil et al.[2003] proposed that some SNC magmas were alkaline.Figure 9 shows the compositions of alkaline (anorogenic)lavas from the Nandewar volcano [Abbott, 1969; Stolz,1985]. Although the Al2O3 and FeO* abundances aresimilar to calc-alkaline lavas, the Mars surface compositionsare clearly distinguished from alkaline rocks by their SiO2

and TiO2 contents.[37] By comparison with terrestrial magmas, it may be

difficult to envision how hydrous melting and fractionalcrystallization could have happened in the absence of platetectonics. Subduction of hydrated slabs provides the mech-anism for adding so much water to the Earth’s mantle. Atpresent, Mars is a one-plate planet that convectively trans-lates radioactive heat in the mantle through a stagnant lid[Schubert et al., 2001]. However, some authors haveproposed an early period of plate tectonics on Mars toexplain various geomorphic features [Sleep, 1994; Baker etal., 2002; Fairen et al., 2002] and magnetic lineationpatterns detected in parts of the ancient southern highlands[Connerney et al., 1999; Nimmo and Stevenson, 2000],although there is little or no supporting evidence [Pruisand Tanaka, 1995]. Thermal models suggest that the stag-nant lid model operating during the entire Martian evolutioncan more readily explain the volume and timing of crustformation than can a model of early plate tectonics followedby a stagnant lid [Hauck and Phillips, 2002; Breuer andSpohn, 2003]. Also, the preservation of early-formed isoto-pic anomalies in the mantle source regions of SNC mete-orites [Chen and Wasserburg, 1986; Borg et al., 1997; Leeand Halliday, 1997] would have been difficult during platetectonic convection.[38] In terrestrial subduction zones, the primary melts are

basaltic, not andesitic, and andesites are produced primarilyby fractional crystallization and assimilation of overlyingsilicic crust [Rudnick, 1995]. As a consequence, andesiticvolcanism on Earth is mostly associated with thick, conti-nental crust. Conversely, on Mars surface type 2 (possiblyandesitic) materials overlie thin crust in the northern plains,whereas the thick southern highlands are overlain by surfacetype 1 (basaltic) materials. Baker et al. [2002] suggestedthat early subduction with attendant hydrous mantle meltingon Mars produced a thick andesitic crust in the highlands.Later floods eroded the highlands andesitic crust anddelivered these sediments to the lowlands, where theycovered thinner oceanic crust. The highlands were subse-quently mantled by younger, plume-related basalts. Thereare several potential problems with this scenario. First, itseems unlikely that large impacts into the highlands wouldnot have excavated significant amounts of andesitic crustalstratigraphy to produce a mixed TES spectral signal. Andsecond, even if surface type 2 is interpreted as weatheredbasalt (see below), our derived composition of surface type 1materials in the highlands still suggests hydrous melting,unlike the tholeiitic compositions expected for plumebasalts and seen in SNC meteorites.[39] It is conceivable that hydrous melting of the

Martian mantle might somehow have occurred withoutplate tectonics. There is precedence for this idea: Rudnick

[1995] argued that the production of a part of the Earth’scontinental crust in the Archaean required a differentmechanism than subduction, although the specific mech-anism is unclear. Lowman [1989] proposed that an ancientandesitic crust could have formed on Mars throughhydrous melting without plate tectonics. In his model,mantle water was thought to be primordial, leading to anearly crust of andesite with later dry, basaltic volcanism.Early melts of a wet mantle would presumably scavengewater, leading to the dry mantle source region of youngerSNC meteorites, as envisioned by Waenke and Dreibus[1988]. Marsh [2002] argued that terrestrial bimodalmagmatism, leading to the production of siliceous meltsfrom basaltic parent magmas, occurs through the forma-tion of lenses within solidification fronts. However, it isnot obvious that such a mechanism could operate on aglobal scale and produce volumetrically significantamounts of andesite.

4.2. Weathering of Basaltic Rocks

[40] Chemical weathering in the terrestrial environmentalmost certainly takes place under conditions different fromMars. However, Nesbitt and Wilson [1992] indicated thatleaching of major elements from weathered volcanic rocksis not greatly influenced by primary mineralogy, bulkchemical composition, or climatic conditions, so it isplausible that tools developed for terrestrial weatheringbut applied to the Martian regolith may provide usefulinsights.[41] Chemical weathering of terrestrial basaltic rocks

typically leads to bulk depletion of leachable CaO, Na2O,K2O, and to a lesser extent MgO, relative to FeO and Al2O3.These depletions are illustrated by arrows in molar ternarydiagrams (Figure 10) devised by Nesbitt and Young [1984]and Nesbitt and Wilson [1992]. The plotted positions of thenew compositions from this study (X-boxes), as well ascompositions based on previous deconvolutions, suggestthat surface type 2 materials could have been produced bychemical weathering of surface type 1 basaltic andesite.Figure 10a shows that surface type 2 materials contain moreAl2O3 than surface type 1, as appropriate for weatheredmaterials, but they also contain slightly higher proportionsof K2O, the component most readily leached. The relativepositions of surface type 1 and 2 materials in Figure 10b arealso consistent with weathering. In Figure 10c, surfacetype 2 compositions are displaced toward higher Al2O3

relative to surface type 1, not directly away from the easilyleached apex. This displacement might occur if iron was notalready oxidized to the ferric state, which is the insolubleform [Nesbitt and Wilson, 1992].[42] Only a modest degree of chemical weathering is

required by these diagrams (the displacement is less thanthe lengths of vectors representing weathered terrestrialbasalts in Figure 10; weathered terrestrial basalts commonlyare shifted across the feldspar-FeO or feldspar-FeO + MgOjoins). Incomplete weathering is in agreement with thelimited proportions of clays and other alteration phases(<50%) in the deconvolved surface type 2 spectra of Wyattand McSween [2002]. Alternatively, surface type 2 materialscould consist of basaltic sand that has been physicallymixed (on TES pixel scale) with chemically weatheredmaterials. The surface type 1 composition might also


Figure 10. Molar variation diagrams illustrating leaching of soluble components during the chemicalweathering of basalt [Nesbitt and Young, 1984; Nesbitt and Wilson, 1992]. Terrestrial weathering trendsare shown by arrows. Surface type 2 compositions generally contain less leachable oxides and could haveformed by chemical weathering of surface type 1 materials. The global dust composition (Table 1) doesnot correspond to either surface type or basaltic shergottites, nor to a simple mixture of any of thesematerials.

Figure 11. Mg/Si versus Al/Si diagram (wt. ratios), commonly used to distinguish igneous rocks fromMars and Earth. Surface type 1 and 2 compositions are displaced from the Martian trend and arrayedalong a gray arrow defined by the compositions of terrestrial palagonitized basalts and a basaltweathering profile [McLennan, 2003]. Data sources as in Figure 2, plus Rieder et al. [1997]. Ratio errorbars based on the relative error analysis of Wyatt et al. [2001].


represent less weathered material, as its TES spectraldeconvolutions utilize smaller amounts of clays [Bandfieldet al., 2000; Hamilton et al., 2001; Wyatt and McSween,2002].[43] Weathering of basalts commonly leads to mobility of

silica. In fact, the silica in basaltic rocks (usually containedin readily altered glass and plagioclase) is more easilymobilized than is the silica in quartz-bearing rocks[McLennan, 2003]. Although silica is commonly leachedfrom basalts during chemical weathering, it ultimatelymust be precipitated and concentrated elsewhere. Indeed,measurements of SiO2/Al2O3 in palagonites and other clayalteration products formed by low-temperature alteration ofIcelandic basaltic glasses show both depletions and enrich-ments of silica [McLennan, 2003]. Some Hawaiian basaltsalso have silica-rich weathering rinds [Farr and Adams,1984; Crisp et al., 1990]. Thus it seems plausible that thehigher silica contents of surface type 2 materials could alsoreflect chemical weathering of basalt.[44] The Mg/Si versus Al/Si diagram (Figure 11) has been

used commonly to distinguish terrestrial and Martian rocks[e.g., Rieder et al., 1997]. Compositions for surface type 1and 2 materials clearly plot off the trend defined by Martianultramafic and basaltic rocks. Instead, they are arrayed alonga trend toward higher Al/Si (gray arrow). This arrow actuallydefines the chemical changes measured in palgonitizedbasalts and in a basalt weathering profile, as summarizedby McLennan [2003]. The positions of surface type 1 and 2compositions and the resulting slope of the trend resemblechemically weathered basalts, not Martian igneous rocks.Surface type 2 is not displaced to higher Al/Si contents aswould be expected if it is more highly weathered; however,mobilization (increase) of silica might have limited Al/Sivariations for surface type 2 and Mars Pathfinder rockcompositions in Figure 11.[45] As noted earlier, the mapped distribution of surface

type 2 in the northern lowlands is concentrated in AcidaliaPlanitia, part of the low-albedo surface of the VastitasBorealis Formation [Bandfield et al., 2000; Wyatt andMcSween, 2002]. This formation was initially mapped onthe basis of its morphologic and albedo characteristics andits occurrence below the Martian highland/lowland bound-ary [Scott et al., 1987]. Its various landforms have beenattributed to volcanic, tectonic, glacial, periglacial, andsedimentary processes [Clifford and Parker, 2001; Headet al., 1999; Kargel et al., 1995; Kreslavsky and Head,2001; McGill and Hills, 1992; Parker et al., 1989; Scott etal., 1987; Tanaka, 1997; Tanaka et al., 2001, 2003]. TheVastitas Borealis plains have most recently been interpretedas sedimentary deposits derived from outside the basin[Head et al., 2002b] or as altered sediments formed throughlocal reworking of earlier deposits by permafrost processes[Tanaka et al., 2003]. The sediments are thought tobe underlain by ancient ridged volcanic plains, possiblyexposed through erosion as basaltic outliers [Rogers andChristensen, 2003]. Both ‘‘new-view’’ theories proposethat these materials have undergone significant reworkingduring transport or by indigenous weathering, so they areconsistent with alteration of basaltic sands within thisdepocenter. Vastitas Borealis morphology is characterizedby a mixture of smooth plains, polygonal troughs andfractures, and pitted domes whose origins appear to involve

interactions with water and/or ice [Head et al., 2002b;Tanaka et al., 2003]. TES spectra of the Vastitas Borealisboundary in southern Acidalia Planitia, where it is notobscured by dust, indicate that the proportion of surfacetype 1 material increases outside the basin [Wyatt et al.,2003].[46] The hypothesis that the ancient Martian crust has

experienced modest chemical weathering and concurrentchemical fractionations may obviate the need to appeal tohydrous melting, especially if both surface types 1 and 2 arepartly weathered to different degrees or mixed with differ-ing amounts of weathered materials. This model may bemore consistent with the conventional view of Mars ashaving a relatively dry mantle and no plate tectonics.[47] Because surface types 1 and 2 materials dominate the

exposed Martian crust, it is logical that one or both mighthave been precursors to the globally homogenized dust.Figure 10 shows that the Martian dust composition (Table 1)is distinct from either surface type. Moreover, the dustcannot be rationalized as a simple mixture of the twosurface types. McSween and Keil [2000] used these dia-grams to argue that the composition of Martian dust couldarise by weathering of basalts having shergottite composi-tions, but not by weathering of the Mars Pathfinder dust-free rock (using a previous calibration of the Pathfinder rockcomposition). None of the rock compositions alone inFigure 10 is an obvious precursor for the recalculated globaldust composition, perhaps suggesting that the process thatproduced the dust was not isochemical. Figure 10c alsoallows the possibility that dust formed by addition of ironoxides to surface types 1 or 2, possibly as a result of aeolianconcentration of dense oxide grains [McLennan, 2000;McSween and Keil, 2000], which would translate thecomposition toward the apex containing FeO. Concentra-tion of iron oxides is also suggested by high and variableFeO/MnO ratios in Mars Pathfinder soils [Foley et al.,2003].

5. Conclusions

[48] We have considered a number of diverse observa-tions to try to resolve the question of whether the Martiancrust contains significant quantities of silicic rock:[49] . With only one exception, Martian meteorites have

young crystallization ages, suggesting that they are samplesof young (dust-covered) volcanic centers such as Tharsis orElysium. On the basis of the compositions of SNCs, therecent crust of Mars is basaltic. ALH84001, the onlymeteoritic sample of the ancient Martian crust, is anultramafic cumulate thought to have formed from basalticmagma.[50] . The absence of major element variations (SiO2,

MgO/Al2O3) that correlate clearly with indications of geo-chemical mixing (radiogenic isotopes, REE fractionations,magma redox state) in basaltic/olivine-phyric shergottitessuggests that any assimilated rock component had a majorelement composition similar to basalt. However, it isunclear whether the admixed component represents theancient crust through which shergottites erupted or a meta-somatised mantle reservoir.[51] . The composition of the APXS-analyzed dust-free

rock surface at the Mars Pathfinder landing site is andesitic,


but it is unclear whether this composition representsvolcanic rock or a silicic weathering rind. New alpha-modeanalyses of excess oxygen in Pathfinder rocks, corres-ponding to �2 wt.% H2O in even highly oxidizedsamples, support the interpretation the rocks have weatheredcoatings.[52] . The analyzed chemistry of uncemented Martian

soils has been used to estimate the composition of fines. Thecomposition of globally homogenized Martian fines couldplausibly mimic the average composition of the exposedcrust. However, the dust composition does not conform tothe compositions of either TES surface type or Martianmeteorites, nor to any simple mixture of these materials.The dust is depleted in soluble oxides and appears to bechemically weathered material.[53] . Previously published deconvolutions of MGS-TES

spectra suggest that basalt/basaltic andesite dominates thesouthern highlands. Spectra for the northern lowlands canbe interpreted as either andesite or partly weathered basalt.The sedimentary origin inferred for the Vastitas BorealisFormation in the northern lowlands is consistent with aweathered origin for this vast exposure of surface type 2material.[54] . Densities for the Martian elastic lithosphere esti-

mated from global topography and gravity are higher thanfor the terrestrial continental crust (having the compositionof andesite) and appear to be consistent with an averagebasaltic composition.[55] This listing does not unambiguously resolve the

question of whether the crust is basaltic or contains asignificant andesitic component. Accordingly, we carefullyconsidered the chemical compositions of crustal materialspreviously calculated from TES spectra for surface type 1and 2, using a variety of spectral end-members. We alsocalculated new compositions on the basis of deconvolvingTES data over a wider spectral range than used in priorstudies.[56] All the derived chemical compositions for surface

type 1 materials correspond to basaltic andesite on the basisof the total alkalis versus silica classification diagram. On aFeO*/MgO versus silica diagram, this calc-alkaline compo-sition resembles magmas generated by fairly high degreesof hydrous melting in terrestrial subduction zones. Thechemical compositions of surface type 2 materials corre-spond to andesite, and could have formed by more exten-sive melting of hydrated mantle rocks with accompanyinghydrous fractionation. This petrogenesis challenges ourunderstanding of anorogenic magmatism, and suggests amuch wetter ancient Martian mantle than is normallysupposed. Early differentiation of the voluminous Martiancrust would have depleted the mantle’s water inventory, sothat subsequent dry melting would produce SNC tholeiiticmagmas.[57] Alternatively, surface type 2 materials could be

partly weathered surface type 1 basaltic andesite on thebasis of chemical weathering diagrams. This model circum-vents the problem of explaining how andesite formed in theabsence of subduction, and is consistent with an interpreta-tion of the Mars Pathfinder dust-free rock composition as aweathering rind. This hypothesis implies that chemicalweathering, at least of limited extent, may be pervasiveon the Martian surface, consistent with inferences of wide-

spread sediments. It is conceivable that surface type 1materials could also be partly weathered (to a lesser degree)or volcanic sand mixed with weathered materials (on a TESpixel scale). Chemical fractionations during weathering and/or admixture of weathered materials could eliminate theneed for a hydrated Martian mantle source for volcanicrocks of the ancient crust.[58] This study refines the nature of both the igneous and

chemical weathering explanations for TES-derived globalsurface units. However, at present, we are left with multipleworking hypotheses for the composition and origin of theMartian crust.

[59] Acknowledgments. Jeff Johnson and Hanna Nekvasil providedinsightful reviews. This work was partly supported by NASA Cosmochem-istry grants NAG5-12896 to HYM and NAG5-10728 to TLG.

ReferencesAbbott, M. J., Petrology of the Nandewar Volcano, N.S.W., Australia,Contrib. Mineral. Petrol., 20, 115–134, 1969.

Ash, R. D., S. F. Knott, and G. Turner, A 4-Ga shock age for a Martianmeteorite and implications for the cratering history of Mars, Nature, 380,57–59, 1996.

Baker, V. R., S. Maruyama, and J. M. Dohm, A theory for the geologicalevolution of Mars and related synthesis (GEOMARS), Lunar Planet. Sci.[CD-ROM], XXXIII, abstract 1586, 2002.

Baloga, S. M., P. J. Mouginis-Mark, and L. S. Glaze, Rheology of a longlava flow at Pavonis Mons, Mars, J. Geophys. Res., 108(E7), 5066,doi:10.1029/2002JE001981, 2003.

Bandfield, J. L., Global mineral distributions on Mars, J. Geophys. Res.,107(E6), 5042, doi:10.1029/2000JE001510, 2002.

Bandfield, J. L., V. E. Hamilton, and P. R. Christensen, A global view ofMartian surface compositions from MGS-TES, Science, 287, 1626–1630, 2000.

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

Bell, J. F., et al., Mineralogic and compositional properties of Martian soiland dust: Results from Mars Pathfinder, J. Geophys. Res., 105, 1721–1755, 2000.

Biemann, K., et al., The search for organic substances and inorganic volatilecompounds on the surface of Mars, J. Geophys. Res., 82, 4641–4658,1977.

Bishop, J. L., S. L. Murchie, C. M. Pieters, and A. P. Zent, A model forformation of dust, soil, and rock coatings on Mars: Physical and chemicalprocesses on the Martian surface, J. Geophys. Res., 107(E11), 5097,doi:10.1029/2001JE001581, 2002.

Blaney, D. L., D. A. Glenar, and G. L. Bjorker, High spectral resolutionspectroscopy of Mars from 2 to 4 mm: Surface mineralogy and the atmo-sphere, in Sixth International Conference on Mars, LPI Contrib. 3237,Lunar and Planet. Inst., Houston, Tex., 2003.

Borg, L. E., L. E. Nyquist, L. A. Taylor, H. Wiesmann, and C.-Y. Shih,Constraints on Martian differentiation processes from Rb-Sr and Sm-Ndisotopic analyses of the basaltic shergottite QUE94201, Geochim. Cos-mochim. Acta, 61, 4915–4931, 1997.

Borg, L. E., L. E. Nyquist, H. Wiesmann, C.-Y. Shih, and Y. Reese, Theage of Dar al Gani 476 and the differentiation history of the Martianmeteorites inferred from their radiogenic isotopic systematics, Geochim.Cosmochim. Acta, 67, 3519–3536, 2003.

Boynton, W. V., et al., Distribution of hydrogen in the near surface ofMars: Evidence for subsurface ice deposits, Science, 297, 81–85,2002.

Breuer, D., and T. Spohn, Early plate tectonics versus single-platetectonics on Mars: Evidence from magnetic field history and crustevolution, J. Geophys. Res., 108(E7), 5072, doi:10.1029/2002JE001999,2003.

Catermole, P., Sequence, rheological properties, and effusion rates ofvolcanic flows at Alba Patera, Mars, Proc. Lunar Planet. Sci. Conf.17th, Part 2, J. Geophys. Res., 92, suppl., E553–E560, 1987.

Chen, J. H., and G. J. Wasserburg, Formation ages and evolution ofShergotty and its parent planet from U-Th-Pb systemics, Geochim. Cos-mochim. Acta, 50, 955–968, 1986.

Christensen, P. R., J. L. Bandfield, M. D. Smith, V. E. Hamilton, and R. N.Clark, Identification of a basaltic component on the Martian surface from


Thermal Emission Spectrometer data, J. Geophys. Res., 105, 9609–9621,2000a.

Christensen, P. R., et al., Detection of crystalline hematite mineralization onMars by the Thermal Emission Spectrometer: Evidence for near-surfacewater, J. Geophys. Res., 105, 9623–9642, 2000b.

Christensen, P. R., J. L. Bandfield, V. E. Hamilton, D. A. Howard, M. D.Lane, J. L. Piatek, S. W. Ruff, and W. L. Stefanov, A thermal emissionspectral library of rock forming minerals, J. Geophys. Res., 105, 9735–9739, 2000c.

Clark, B. C., A. K. Baird, R. J. Weldon, D. M. Tsuasaki, L. Schabel, andM. P. Candelaria, Chemical composition of Martian fines, J. Geophys.Res., 87, 10,059–10,067, 1982.

Clifford, S. M., and T. J. Parker, The evolution of the Martian hydrosphere:Implications for the fate of a primordial ocean and the current state of thenorthern plains, Icarus, 154, 40–79, 2001.

Connerney, J. E. P., M. H. Acuna, P. Wasilewski, N. F. Ness, H. Reme,C. Mazelle, D. Vignes, R. P. Lin, D. Mitchell, and P. Cloutier, Mag-netic lineations in the ancient crust of Mars, Science, 284, 794–798,1999.

Crisp, J. A., A. B. Kahle, and E. A. Abbott, Thermal infrared spectralcharacter of Hawaiian basaltic glasses, J. Geophys. Res., 95, 21,657–21,669, 1990.

Dreibus, G., B. Spettel, R. Haubold, K. P. Jochum, H. Palme, D. Wolf,and J. Zipfel, Chemistry of a new shergottite: Sayh Al Uhaymir 005,Meteorit. Planet. Sci., 35, A49, 2000.

Fairen, A. G., J. Ruiz, and F. Anguita, An origin for the linear magneticanomalies on Mars through accretion of terranes: Implications fordynamo timing, Icarus, 160, 220–223, 2002.

Farr, T. G., and J. B. Adams, Rock coatings in Hawaii, Geol. Soc. Am. Bull.,95, 1077–1083, 1984.

Feeley, K. C., and P. R. Christensen, Quantitative compositional analysisusing thermal emission spectroscopy: Application to igneous and meta-morphic rocks, J. Geophys. Res., 104, 24,195–24,210, 1999.

Folco, L., I. A. Franchi, M. D’Orazio, S. Rocchi, and L. Schultz, A newMartian meteorite from the Sahara: The shergottite Dar al Gani 489,Meteorit. Planet. Sci., 35, 827–839, 2000.

Foley, C. N., T. E. Economou, R. N. Clayton, and W. Dietrich, Calibrationof the Mars Pathfinder alpha proton X-ray spectrometer, J. Geophys.Res., 108(E12), 8095, doi:10.1029/2002JE002018, 2003.

Frey, H. V., J. H. Roark, K. M. Shockey, E. L. Frey, and S. E. H. Sakimoto,Ancient lowlands on Mars, Geophys. Res. Lett., 29(10), 1384,doi:10.1029/2001GL013832, 2002.

Gaetani, G. A., and T. L. Grove, The influence of water on melting ofmantle peridotite, Contrib. Mineral. Petrol., 131, 323–346, 1998.

Greeley, R., and P. D. Spudis, Volcanism on Mars, Rev, Geophys., 19, 13–41, 1981.

Grove, T. L., R. J. Kinzler, and W. B. Bryan, Fractionation of mid-oceanridge basalt (MORB), in Mantle Flow and Melt Migration at Mid-OceanRidges, Geophys. Monogr. Ser., vol. 71, edited by J. Phipps-Morgan etal., pp. 281–311, AGU, Washington, D. C., 1993.

Grove, T. L., S. W. Parman, S. A. Bowring, R. C. Price, and M. B. Baker,The role of an H2O-rich fluid component in the generation of primitivebasaltic andesites and andesites from the Mt. Shasta region, N. California,Contrib. Mineral. Petrol., 142, 375–396, 2002.

Grove, T. L., L. T. Elkins-Tanton, S. W. Parman, N. Chatterjee, O. Muntener,and G. A. Gaetani, Fractional crystallization and mantle melting controlson calc-alkaline differentiation trends, Contrib. Mineral. Petrol., 145,515–533, 2003.

Hale, V. P. S., H. Y. McSween, and G. A. McKay, Re-evaluation ofintercumulus liquid composition and oxidation state for the Shergottymeteorite, Geochim. Cosmochim. Acta, 63, 1459–1470, 1999.

Hamilton, V. E., and P. R. Christensen, Determining the modal mineralogyof mafic and ultramafic igneous rocks using thermal emission spectro-scopy, J. Geophys. Res., 105, 9717–9733, 2000.

Hamilton, V. E., M. B. Wyatt, H. Y. McSween, and P. R. Christensen,Analysis of terrestrial and Martian volcanic compositions using thermalemission spectroscopy: 2. Application to Martian surface spectra fromMGS TES, J. Geophys. Res., 106, 14,733–14,746, 2001.

Hamilton, V. E., P. R. Christensen, H. Y. McSween, and J. L. Bandfield,Searching for the source regions of Martian meteorites using MGS TES:Integrating Martian meteorites into the global distribution of igneousmaterials on Mars, Meteorit. Planet. Sci., 38, 871–886, 2003.

Harper, C. L., L. E. Nyquist, B. M. Bansal, H. Wiesmann, and C.-Y. Shih,Rapid accretion and early differentiation of Mars indicated by142Nd/144Nd in SNC meteorites, Science, 267, 213–216, 1995.

Hartmann, W. K., and G. Neukum, Cratering chronology and the evolutionof Mars, Space Sci. Rev., 96, 165–194, 2001.

Hauck, S. A., and R. J. Phillips, Thermal and crustal evolution ofMars, J. Geophys. Res., 107(E7), 5052, doi:10.1029/2001JE001801,2002.

Head, J. W., H. Hiesinger, M. A. Ivanov, M. A. Kreslavsky, S. Pratt, andB. J. Thomson, Possible ancient oceans on Mars: New evidence from theMars Orbiter Laser Altimeter, Science, 286, 2134–2137, 1999.

Head, J. N., H. J. Melosh, and B. A. Ivanov, Martian meteorite launch:High-speed ejecta from small craters, Science, 298, 1752–1756,2002a.

Head, J. W., M. A. Kreslavsky, and S. Pratt, Northern lowlands of Mars:Evidence for widespread volcanic flooding and tectonic deformation inthe Hesperian Period, J. Geophys. Res., 107(E1), 5003, doi:10.1029/2000JE001445, 2002b.

Herd, C. D. K., L. E. Borg, J. H. Jones, and J. J. Papike, Oxygen fugacityand geochemical variations in the Martian basalts: Implications forMartian basalt petrogenesis and the oxidation state of the upper mantleof Mars, Geochim. Cosmochim. Acta, 66, 2025–2036, 2002.

Hooft, E. E., and R. S. Detrick, The role of density in the accumulation ofbasaltic melts at mid-ocean ridges, Geophys. Res. Lett., 20, 423–426,1993.

Imae, N., Y. Ikeda, K. Shinoda, H. Kojima, and N. Iwata, Yamato nakhlites:Petrography and mineralogy, Antarct. Meteorite Res., 16, 13–33,2003.

Johnson, J. R., et al., Preliminary results on photometric properties ofmaterials at the Sagan Memorial Station, Mars, J. Geophys. Res., 104,8809–8830, 1999.

Johnson, J. R., F. Horz, P. G. Lucey, and P. R. Christensen, Thermal infra-red spectroscopy of experimentally shocked anorthosite and pyroxenite:Implications for remote sensing of Mars, J. Geophys. Res., 107(E10),5073, doi:10.1029/2001JE001517, 2002.

Juster, T. C., T. L. Grove, and M. J. Perfit, Experimental constraints on thegeneration of FeTi basalts, andesites and rhyodacites at the GalapagosSpreading Center, 85�W and 95�W, J. Geophys. Res., 94, 9251–9274,1989.

Kargel, J. S., V. R. Baker, J. E. Beget, J. F. Lockwood, T. L. Pewe, J. S.Shaw, and R. G. Strom, Evidence of ancient continental glaciation in theMartian northern plains, J. Geophys. Res., 100, 5351–5368, 1995.

Kinzler, R. J., J. M. Donnelly-Nolan, and T. L. Grove, Late Holocenehydrous mafic magmatism at the Paint Pot Crater and Callahan flows,Medicine Lake volcano, N. California and the influence of H2O in thegeneration of silicic magmas, Contrib. Mineral. Petrol., 138, 1 –16,2000.

Kraft, M. D., T. G. Sharp, and J. R. Michalski, Thermal emission spectra ofsilica-coated basalt and considerations for Martian surface mineralogy,Lunar Planet. Sci. [CD-ROM], XXXIV, abstract 1420, 2003.

Kreslavsky, M. A., and J. W. Head, Stealth craters in the northern lowlandsof Mars: Evidence for a buried Early-Hesperian-aged unit, Lunar Planet.Sci. [CD-ROM], XXXII, abstract 1001, 2001.

Larsen, K. W., R. E. Arvidson, B. L. Jolliff, and B. C. Clark, Correspon-dence and least squares analyses of soil and rock compositions for theViking Lander 1 and Pathfinder landing sites, J. Geophys. Res., 105,29,207–29,221, 2000.

Le Bas, M. J., R. W. Le Maitre, A. Streckeisen, and B. Zanettin, A chemicalclassification of volcanic rocks based on the total alkali-silica diagram,J. Petrol., 27, 745–750, 1986.

Lee, D.-C., and A. N. Halliday, Core formation on Mars and differentiatedasteroids, Nature, 388, 854–857, 1997.

Lodders, K., A survey of shergottite, nakhlite and chassigny meteoriteswhole-rock compositions, Meteorit. Planet. Sci., 22, A183–A190, 1998.

Lowman, P. D., Jr., Comparative planetology and the origin of continentalcrust, Precambrian Res., 44, 171–195, 1989.

Malin, M. C., and K. S. Edgett, Sedimentary rocks of early Mars, Science,290, 1927–1937, 2000.

Malin, M. C., and K. S. Edgett, Mars Global Surveyor Mars OrbiterCamera: Interplanetary cruise through primary mission, J. Geophys.Res., 106, 23,429–23,570, 2001.

Marsh, B. D., On bimodal differentiation by solidification front instabilityin basaltic magmas, part 1: Basic mechanics, Geochim. Cosmochim. Acta,66, 2211–2229, 2002.

McCord, T. B., R. B. Singer, B. R. Hawke, J. B. Adams, D. L. Evans, J. W.Head, P. J. Mouginis-Mark, C. M. Pieters, R. L. Huguenin, and S. H.Zisk, Mars: Definition and characterization of global surface units withemphasis on composition, J. Geophys. Res., 87, 10,129–10,148, 1982.

McEwen, A. S., M. C. Malin, M. H. Carr, and W. K. Hartmann, Volumi-nous volcanism on early Mars revealed in Valles Marineris, Nature, 397,584–586, 1999.

McGill, G. E., and L. S. Hills, Origin of giant Martian polygons, J. Geo-phys. Res., 97, 2633–2647, 1992.

McGovern, P. J., S. C. Solomon, D. E. Smith, M. T. Zuber, M. Simons,M. A. Wiezorek, R. J. Phillips, G. A. Neumann, O. Aharonson, and J. W.Head, Localized gravity/topography admittance and correlation spectraon Mars: Implications for regional and global evolution, J. Geophys.Res., 107(E12), 5136, doi:10.1029/2002JE001854, 2002.


McKenzie, D. D., D. Barnett, and D.-N. Yuan, The relationship betweenMartian gravity and topography, Earth Planet. Sci. Lett., 5, 1–16,2002.

McLennan, S. M., Chemical composition of Martian soil and rocks: Com-plex mixing and sedimentary transport, Geophys. Res. Lett., 27, 1335–1338, 2000.

McLennan, S. M., Sedimentary silica on Mars, Geology, 31, 315–318,2003.

McLennan, S. M., and S. R. Taylor, Archaean sedimentary rocks andtheir relation to the composition of the Archaean continental crust, inArchaean Geochemistry, edited by A. Kroner et al., pp. 47–72, Spring-er-Verlag, New York, 1984.

McSween, H. Y., The rocks of Mars, from far and near, Meteorit. Planet.Sci., 37, 7–25, 2002.

McSween, H. Y., and K. Keil, Mixing relationships in the Martian regolithand the composition of globally homogenous dust, Geochim. Cosmo-chim. Acta, 64, 2155–2166, 2000.

McSween, H. Y., et al., Chemical, multispectral and textural constraintson the composition and origin of rocks at the Mars Pathfinder landingsite, J. Geophys. Res., 104, 8679–8715, 1999.

Michael, P. J., and W. C. Cornell, Influence of spreading rate and magmasupply on crystallization and assimilation beneath mid-ocean ridges:Evidence from chlorine and major element chemistry of mid-ocean ridgebasalts, J. Geophys. Res., 103, 18,235–18,256, 1998.

Mittlefehldt, D. W., ALH84001, a cumulate orthopyroxenite member of theMartian meteorite clan, Meteoritics, 29, 214–221, 1994.

Miyashiro, A., Volcanic rock series in island arc and active continentalmargins, Am. J. Sci., 274, 321–355, 1974.

Morris, R. V., et al., Mineralogy, composition, and alteration of MarsPathfinder rocks and soils: Evidence from multispectral, elemental, andmagnetic data on terrestrial analogue, SNC meteorite, and Pathfindersamples, J. Geophys. Res., 105, 1757–1817, 2000.

Morris, R. V., T. G. Graff, S. A. Mertzman, M. D. Lane, and P. R.Christensen, Palagonitic (not andesitic) Mars: Evidence from ThermalEmission and VNIR spectra of palagonitic alteration rinds on basalticrocks, in Sixth International Conference on Mars, LPI Contrib. 3211,Lunar and Planet. Inst., Houston, Tex., 2003.

Muentener, O., P. B. Kelemen, and T. L. Grove, The role of H2O on thecrystallization of primitive arc magmas at uppermost mantle conditionsand genesis of igneous pyroxenites: An experimental study, Contrib.Mineral. Petrol., 141, 643–658, 2000.

Murchie, S., L. Kirkland, S. Erard, J. Mustard, and M. Robinson, Near-infrared spectral variations of Martian surface materials from ISMimaging spectrometer data, Icarus, 147, 444–471, 2000.

Mustard, J. F., S. Murchie, S. Erard, and J. M. Sunshine, In situ composi-tions of Martian volcanics: Implications for the mantle, J. Geophys. Res.,102, 25,605–25,615, 1997.

Nekvasil, H., J. Filiberto, M. Whitacker, D. H. Lindsley, Magmas parentalto the Chassigny meteorite: New considerations, in Sixth InternationalConference on Mars [CD-ROM], LPI Contrib. 3041, Lunar and Planet.Inst., Houston, Tex., 2003.

Nesbitt, H. W., and G. M. Young, Prediction of some weathering trends ofplutonic and volcanic rocks based on thermodynamic and kinetic con-siderations, Geochim. Cosmochim. Acta, 48, 1523–1534, 1984.

Nesbitt, H. W., and R. E. Wilson, Recent chemical weathering of basalts,Am. J. Sci., 292, 740–777, 1992.

Nimmo, F., Admittance estimates of mean crustal thickness and density atthe Martian hemispheric dichotomy, J. Geophys. Res., 107(E11), 5117,doi:10.1029/2000JE001488, 2002.

Nimmo, F., and D. Stevenson, Influence of early plate tectonics on thethermal evolution and magnetic field of Mars, J. Geophys. Res., 105,11,969–11,979, 2000.

Nyquist, L. E., D. D. Bogard, C.-Y. Shih, A. Greshake, D. Stoffler, andO. Eugster, Ages and geologic histories of Martian meteorites, Space Sci.Rev., 96, 105–164, 2001.

Parker, T. J., R. S. Saunders, and D. M. Schneeberger, Transitional mor-phology in west Deuteronilus Mensae, Mars: Implications for modifica-tion of the lowland/upland boundary, Icarus, 82, 111–145, 1989.

Phillips, R. J., et al., Ancient geodynamics and global-scale hydrology onMars, Science, 291, 2587–2591, 2001.

Pruis, M. J., and K. L. Tanaka, The Martian northern plains did not resultfrom plate tectonics, Lunar Planet Sci., 1147–1148, 1995.

Ramsey, M. S., and P. R. Christensen, Mineral abundance determination:Quantitative deconvolution of thermal emission spectra, J. Geophys. Res.,103, 577–596, 1998.

Rieder, R., T. Economou, H. Wanke, A. Turkevich, J. Crisp, J. Bruckner,G. Dreibus, and H. Y. McSween, The chemical composition of Martiansoil and rocks returned by the mobile alpha proton X-ray spectrometer:Preliminary results from the X-ray mode, Science, 278, 1771–1774,1997.

Rogers, D., and P. R. Christensen, Age relationship of basaltic and andesiticsurface compositions on Mars: Analysis of high-resolution TES observa-tions of the northern hemisphere, J. Geophys. Res., 108(E4), 5030,doi:10.1029/2002JE001913, 2003.

Rubin, A. E., P. H. Warren, J. P. Greenwood, R. S. Verish, L. A. Leshin,R. L. Hervig, R. N. Clayton, and T. K. Mayeda, Los Angeles: The mostdifferentiated basaltic Martian meteorite, Geology, 28, 1011–1014,2000.

Rudnick, R. L., Making continental crust, Nature, 278, 571–578, 1995.Ruff, S. W., Basaltic andesite or weathered basalt: A new assessment, inSixth International Conference on Mars, LPI Contrib. 3258, Lunar andPlanet. Inst., Houston, Tex., 2003.

Ruff, S. W., and P. R. Christensen, Identifying compositional heterogeneityin Mars’ Nili Patera caldera using THEMIS and TES data, Lunar Planet.Sci. [CD-ROM], XXXXIV, abstract 2068, 2003.

Ryan, M. P., Neutral buoyancy and the structure of mid-ocean ridge magmareservoirs, J. Geophys. Res., 98, 22,321–22,338, 1993.

Schubert, G., D. L. Turcotte, and P. Olsen, Mantle Convection in theEarth and Planets, 940 pp., Cambridge Univ. Press, New York,2001.

Scott, D. H., K. L. Tanaka, R. Greeley, and J. E. Guest, Geologic maps ofthe western and eastern equatorial and polar regions of Mars, U.S. Geol.Surv. Misc. Invest. Ser., Map I-1802–A, B, C, 1987.

Sisson, T. W., and T. L. Grove, Experimental investigations of the role ofwater in calc-alkaline differentiation and subduction zone magmatism,Contrib. Mineral. Petrol., 113, 143–166, 1993.

Sleep, N. H., Martian plate tectonics, J. Geophys. Res., 99, 5639–5655,1994.

Smith, M. D., J. L. Bandfield, and P. R. Christensen, Separation of atmo-spheric and surface spectral features in Mars Global Surveyor ThermalEmission Spectrometer (TES) spectra, J. Geophys. Res., 105, 9589–9607, 2000.

Solomon, S. C., and J. W. Head, Evolution of the Tharsis province of Mars:The importance of heterogeneous lithospheric thickness and volcanicconstruction, J. Geophys. Res., 82, 9755–9774, 1982.

Stolper, E., and D. Walker, Melt density and the average composition ofbasalt, Contrib. Mineral. Petrol., 74, 7–12, 1980.

Stolz, A. J., The role of fractional crystallization in the evolution of theNandewar Volcano, northeastern New South Wales, Australia, J. Petrol.,26, 1002–1026, 1985.

Tanaka, K. L., Sedimentary history and mass flow structures of Chryse andAcidalia Planitiae, Mars, J. Geophys. Res., 102, 4131–4150, 1997.

Tanaka, K. L., N. K. Isbell, D. H. Scott, R. Greeley, and J. E. Guest,The resurfacing history of Mars: A synthesis of digitized, Viking-based geology, Proc. Lunar Planet. Sci. Conf. 18th, 665 –678,1988.

Tanaka, K. L., W. B. Banerdt, J. S. Kargel, and N. Hoffman, Huge, CO2–charged debris-flow deposit and tectonic sagging in the northern plains ofMars, Geology, 27, 427–430, 2001.

Tanaka, K. L., J. A. Skinner Jr., T. M. Hare, T. Joyal, and A. Wenker,Resurfacing history of the northern plains of Mars based on geologicmapping of Mars Global Surveyor data, J. Geophys. Res., 108(E4),8043, doi:10.1029/2002JE001908, 2003.

Toulmin, P., III, et al., Geochemical and mineralogical interpretations of theViking inorganic chemical results, J. Geophys. Res., 82, 4625–4634,1977.

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

Turcotte, D. L., R. Shcherbakov, B. D. Malamud, and A. B. Kucinkas, Isthe Martian crust also the Martian elastic lithosphere?, J. Geophys. Res.,107(E11), 5091, doi:10:1029/2001JE001594, 2002.

Wadhwa, M., Redox state of Mars’ upper mantle and crust from Eu anoma-lies in shergottite pyroxenes, Science, 291, 1527–1530, 2001.

Wadhwa, M., and T. L. Grove, Archean cratons on Mars? Evidencefrom trace elements, isotopes and oxidation states of SNC meteorites,Geochim. Cosmochim. Acta, 66, A816, 2002.

Waenke, H., and G. Dreibus, Chemical composition and accretion historyof terrestrial planets, Philos. Trans. R. Soc. London, Ser. A, 325, 545–557, 1988.

Waenke, H., J. Bruckner, G. Dreibus, R. Rieder, and I. Ryabchikov,Chemical composition of rocks and soils at the Pathfinder site, SpaceSci. Rev., 96, 317–330, 2001.

Wyatt, M. B., and H. Y. McSween, Spectral evidence for weathered basaltas an alternative to andesite in the northern lowlands of Mars, Nature,417, 263–266, 2002.

Wyatt, M. B., V. E. Hamilton, H. Y. McSween, P. R. Christensen, and L. A.Taylor, Analysis of terrestrial and Martian volcanic compositions usingthermal emission spectroscopy: 1. Determination of mineralogy, chemis-try, and classification strategies, J. Geophys. Res., 106, 14,711–14,732,2001.


Wyatt, M. B., H. Y. McSween, J. E. Moersch, and P. R. Christensen,Analysis of surface compositions in the Oxia Palus region on Marsfrom Mars Global Surveyor Thermal Emission Spectrometer observa-tions, J. Geophys. Res., 108(E9), 5107, doi:10.1029/2002JE001986,2003.

Yen, A. S., B. C. Murray, and G. R. Rossman, Water content of the Martiansoil: Laboratory simulations of reflectance spectra, J. Geophys. Res., 103,11,125–11,133, 1998.

Zimbelman, J. R., Estimates for rheological properties of flows on theMartian volcano Ascraeus Mons, Proc. Lunar Planet. Sci. Conf. 16th,Part 1, J. Geophys. Res., 90, suppl., D157–D162, 1985.

Zuber, M. T., The crust and mantle of Mars, Nature, 412, 220–227,2001.

Zuber, M. T., et al., Internal structure and early thermal evolution ofMars from Mars Global Surveyor topography and gravity, Science,287, 1788–1793, 2001.

��T. L. Grove, Department of Earth, Atmospheric and Planetary Sciences,

Massachusetts Institute of Technology, Cambridge, MA 02139, USA.([email protected])H. Y. McSween Jr., Department of Earth and Planetary Sciences,

University of Tennessee, Knoxville, TN 37996-1410, USA. ([email protected])M. B. Wyatt, Department of Geology, Arizona State University, Tempe,

AZ 85287-1404, USA. ([email protected])


Date post:	04-Jul-2020
Category:	Documents
Upload:	others
View:	1 times
Download:	0 times

Constraints on the composition and petrogenesis of the ...these compositions might have arisen on...

Documents