Elsevier

Earth and Planetary Science Letters

Volume 410, 15 January 2015, Pages 140-151
Earth and Planetary Science Letters

Meteoritic evidence for a previously unrecognized hydrogen reservoir on Mars

https://doi.org/10.1016/j.epsl.2014.11.022Get rights and content

Highlights

  • We provide evidence for a previously unrecognized hydrogen reservoir on Mars.

  • This reservoir has a non-atmospheric hydrogen isotopic composition.

  • This hydrogen reservoir represents hydrated crust and/or ground ice.

  • This reservoir could account for a large part of the initial water budge of Mars.

Abstract

Fluvial landforms on Mars suggest that it was once warm enough to maintain persistent liquid water on its surface. The transition to the present cold and dry Mars is closely linked to the history of surface water, yet the evolution of surficial water is poorly constrained. Based on in situ hydrogen isotope (D/H) analyses of quenched and impact glasses in Martian meteorites, we provide evidence for the existence of a distinct but ubiquitous water/ice reservoir (D/H=23 times Earth's ocean water) that lasted from at least the time when the meteorites crystallized (173–472 million years ago) to the time they were ejected by impacts (0.7–3.3 million years ago), but possibly much longer. The origin of this reservoir appears to predate the current Martian atmospheric water (D/H=56 times Earth's ocean water) and is unlikely to be a simple mixture of atmospheric and primordial water retained in the Martian mantle (D/H ≈ Earth's ocean water). This reservoir could represent hydrated crust and/or ground ice interbedded within sediments. Our results corroborate the hypothesis that a buried cryosphere accounts for a large part of the initial water budget of Mars.

Introduction

Recent Mars exploration missions have provided compelling evidence for the presence of liquid water and warmer climatic conditions during the earliest geologic era (Noachian: ∼3.9 to 4.5 Ga) of Mars. Observations of dense valley networks and evaporites (e.g., gypsum) and hydrous minerals (e.g., clays) that are commonly formed by aqueous processes imply that some time in the past Mars had an Earth-like hydrological cycle with large lakes or oceans (Carr, 2006). In contrast to this ancient warm environment, today the surface of Mars is cold and dry. The recent desert-like surface conditions, however, do not necessarily indicate a lack of surface or near-surface water/ice. Massive deposits of subsurface ice have been detected using subsurface radar sounder observations by the European Mars Express mission (Mouginot et al., 2012). Furthermore, landforms that appear to be glacial commonly occur even in the equatorial regions; this has led to the hypothesis that, like on Earth, Mars had recent ice ages in which water–ice would have globally covered the surface (Head et al., 2003, Head et al., 2005). The amount and physical state (ice, liquid, and vapor) of surficial water are closely linked to climate and the potential for life on Mars. Hence, it is important to constrain the global inventory of surficial water/ice reservoirs on Mars and how they have changed with time.

The evolution of surface water/ice and its interaction with the atmosphere can be traced by measurements of hydrogen isotope ratios (D/H: deuterium/hydrogen). Hydrogen is a major component of water (H2O) and its isotopes fractionate significantly during hydrological cycling between the atmosphere, surface waters, ground ice, and the polar ice caps. Telescopic studies have reported that the hemispheric mean D/H ratio of the Martian atmosphere is ∼6 times the terrestrial values (δD5000: Owen et al., 1988); δD=[(D/H)sample/(D/H)reference1]×1000, where the reference is Standard Mean Ocean Water (SMOW). The high atmospheric D/H ratio has been interpreted to result from the preferential loss of hydrogen relative to the heavier deuterium from the Martian atmosphere throughout the planet's history (Bibring et al., 2006, Jakosky and Phillips, 2001, Kurokawa et al., 2014). Recent telescopic measurements have revealed a strong latitudinal variation in the hydrogen isotopic composition of atmospheric water, ranging from low δD (3000) near the polar regions to high δD (6000) near the equatorial regions (Novak et al., 2011). This latitudinal D/H variation is explained by the interplay between multi-latitudinal reservoirs with different average D/H ratios or different latitudinal water/ice abundances (Fisher et al., 2008, Fisher, 2007). While telescopic measurements help constrain the hydrogen isotopic compositions of present-day surficial water reservoirs, subsurface water–ice and the historical variations in the isotopic composition(s) of surface or near-surface water/ice reservoirs are potentially contained in Martian meteorites that formed at different times over the past 0.15 to 4.1 Ga (Lapen et al., 2010, McSween, 2007, Nyquist et al., 2009).

Martian meteorites are all igneous rocks that were either extruded onto the surface as lava flows or solidified as magma bodies in the shallow crust (McSween, 2007), except for the recently found Martian breccia NWA 7034/7533 (Agee et al., 2013, Humayun et al., 2013). Hydrogen isotope analyses of these Martian meteorites provide a significantly greater δD range (111 to +6034, Boctor et al., 2003, Greenwood et al., 2008, Hallis et al., 2012a, Hallis et al., 2012b, Hu et al., 2014, Leshin, 2000, Usui et al., 2012, Watson et al., 1994) than the δD range for most terrestrial water (350 to +50: Hoefs, 2009). Such variation has been interpreted to result from the mixing of two distinct components: a D-enriched component derived from crustal fluids that had interacted with the Martian atmosphere (δD=5000), and a D-poor component that is from the Martian mantle and/or terrestrial contamination. The D/H ratio of the D-poor, primordial mantle component has not been well constrained, ranging from 111 (Hallis et al., 2012a) to +900 (Leshin, 2000). Such uncertainty has arisen both from the scarcity of primitive Martian meteorites and as a result of secondary processes (including contamination by terrestrial water) that obscure the signature of the Martian mantle. The most promising estimate for the Martian primordial water (δD275) is obtained from olivine-hosted melt inclusions in the most magnesian olivine-phyric shergottite Yamato 980459 (Usui et al., 2012).

Martian meteorites contain hydrogen-bearing magmatic phases such as apatite (e.g., Boctor et al., 2003, Filiberto and Treiman, 2009, Hallis et al., 2012a, McCubbin et al., 2012) and water-bearing melt inclusions trapped in nominally anhydrous minerals (e.g., olivine) (e.g., Boctor et al., 2003, Usui et al., 2012). Analyses of these magmatic phases provide the hydrogen isotopic compositions of water in their source magmas that originated from the deep crust and mantle (Boctor et al., 2003, Greenwood et al., 2008, Hallis et al., 2012a, Usui et al., 2012). On the other hand, some Martian meteorites contain post-magmatic glass formed either during volcanic eruption or by the impact events that ejected their host meteorites from Mars and delivered them to Earth. These impact melts are known to have trapped surficial and atmospheric components (Bogard and Johnson, 1983, Rao et al., 2011, Treiman et al., 2000). Thus, comprehensive hydrogen isotope analyses of residual volcanic glass and impact melt-glass can potentially reveal the history of Martian surface water over the last 4Ga.

Hydrogen isotope measurements have been mostly obtained from apatite because it contains high water abundances (5000ppm) and is commonly found in Martian meteorites (Boctor et al., 2003, Greenwood et al., 2008, Hallis et al., 2012a, Leshin, 2000); melt inclusion measurements are less common, due in part to their scarcity (Boctor et al., 2003, Usui et al., 2012). Despite the fact that several Martian meteorites contain ample glass in their matrix, only a few measurements have been reported (e.g., Boctor et al., 2003). The absence of D/H impact melt data mainly reflects the analytical difficulties posed by their low water contents and potential terrestrial contamination.

This study reports hydrogen isotope compositions (δD) and volatile abundances (H2O, CO2, F, Cl, S) of matrix glass phases in Martian meteorites. A recently developed sample preparation method (Usui et al., 2012) enables low-volatile contamination, in situ ion microprobe analyses. Three Martian basaltic meteorites (olivine-phyric shergottites): Yamato 980459 (Y98), Elephant Moraine 79001 (EETA79), and Larkman Nunatak 06319 (LAR06) were studied. These shergottites are geochemically distinct in terms of their radiogenic isotopes (e.g., Sr, Nd and Hf) and relative trace element abundances (e.g., rare earth elements), suggesting that they originated from different magmatic sources (see Section 3). They also have different crystallization and ejection ages, indicating different magmatic events and launching sites from Mars, respectively. These lines of evidence suggest that δD values of matrix/impact melt glasses from the three meteorites potentially provide geochemical information of surficial water–ice reservoirs from different localities and different times on Mars. Remarkably, in situ ion microprobe analyses of these geochemically different shergottites (this study) demonstrate that Mars may have a ubiquitous hydrogen reservoir with an intermediate D/H ratio (δD=10002000) that has previously gone unrecognized.

Section snippets

Low-contamination sample preparation

Hydrogen contamination of thin section samples during preparation, storage, and during the ion microprobe analysis can lead to analytical artifacts as shown by Boctor et al. (2003), Greenwood et al. (2008). Sources of these contaminants include air left in the vacuum system, oils (and/or water) used as lubricants during polishing, and epoxy (or acrylic) resin used as a mounting agent. Among them, resins are the most unavoidable blank sources for Martian meteorites that were highly shocked

Samples

The three olivine-phyric shergottites selected in this study originated from different magma sources in terms of their radiogenic isotopes (e.g., Sr, Nd and Hf) and incompatible trace element compositions (Nyquist et al., 2001, Shafer et al., 2010, Shih et al., 2005). Y98, EETA79, and LAR06 represent geochemically depleted, intermediate, and enriched reservoirs, respectively. These samples also have different crystallization ages (472 Ma for Y98, 173 Ma for EETA79, 193 Ma for LAR06) and Mars

Major element compositions

Only one impact melt pocket was found in the studied sections of Lithology-A of EETA79, whereas four of them were found in the LAR06 sections. The impact melts are characterized by high Al2O3 (20wt.%) and Na2O (3wt.%) contents, as well as moderate mafic mineral components (e.g., 510wt.% FeO and 0.40.8wt.% TiO2) (Table 3) (Fig. 2). This suggests that the impact melts reflect melting of primary plagioclase and pyroxene. Elevated and variable P2O5 contents indicate minor (<1%) contributions

Effect of terrestrial contamination during hydrogen isotope measurements

The influence of terrestrial contamination during the ion microprobe analyses on the hydrogen isotope compositions of Y98 groundmass glasses were already examined in our previous paper (Usui et al., 2012) and this paper reports only “un-contaminated” data. Contributions to the analyses from contamination on grain boundaries and in micro-fractures within the samples can be minimized by checking of the H1, C12, 30Si, and 27Al16O secondary ion images (Section 2.3). Contamination can be also

Conclusions

This study conducted low-contamination ion microprobe analyses of hydrogen isotope compositions (δD) and volatile abundances (H2O, CO2, F, Cl, S) of matrix glass phases in the three geochemically different olivine-phyric shergottites: Y98 (depleted), EETA79 (intermediate), and LAR06 (enriched). These glassy phases in the three shergottites individually exhibit linear negative trends in a δD vs. 1/H2O mixing diagram. These negative trends are explained by two-component mixing between Martian

Acknowledgments

The NASA Antarctic Meteorite Collection and Curation group and the National Institute of Polar Research, Japan are thanked for providing the shergottite samples. We are grateful to B.M. Jakosky, L. Hallis, and an anonymous reviewer for constructive reviews, and T.M. Harrison for editorial handling. Finally, TU thanks H. Kurokawa, M. Sato, J.M. Dohm, and the members of Mars Science Team of Tokyo Tech. for insightful discussions for this project. This work was supported by a NASA Mars Fundamental

References (75)

  • L. Hallis et al.

    Hydrogen isotope analyses of alteration phases in the nakhlite Martian meteorites

    Geochim. Cosmochim. Acta

    (2012)
  • E. Hauri et al.

    SIMS analysis of volatiles in silicate glasses 1. Calibration, matrix effects and comparisons with FTIR

    Chem. Geol.

    (2002)
  • E.H. Hauri et al.

    Matrix effects in hydrogen isotope analysis of silicate glasses by SIMS

    Chem. Geol.

    (2006)
  • C.D.K. Herd et al.

    Oxygen fugacity and geochemical variations in the Martian basalts: implications for Martian basalt petrogenesis and the oxidation state of the upper mantle of Mars

    Geochim. Cosmochim. Acta

    (2002)
  • S. Hu et al.

    NanoSIMS analyses of apatite and melt inclusions in the GRV 020090 Martian meteorite: hydrogen isotope evidence for recent past underground hydrothermal activity on Mars

    Geochim. Cosmochim. Acta

    (2014)
  • M.A. Kendrick et al.

    Contrasting behaviours of CO2, S, H2O and halogens (F, Cl, Br, and I) in enriched-mantle melts from Pitcairn and Society seamounts

    Chem. Geol.

    (2014)
  • V.A. Krasnopolsky et al.

    Oxygen and carbon isotope ratios in the Martian atmosphere

    Icarus

    (2007)
  • H. Kurokawa et al.

    Evolution of water reservoirs on Mars: constraints from hydrogen isotopes in Martian meteorites

    Earth Planet. Sci. Lett.

    (2014)
  • N.P. Lang et al.

    Tharsis-sourced relatively dust-free lavas and their possible relationship to Martian meteorites

    J. Volcanol. Geotherm. Res.

    (2009)
  • L.A. Leshin et al.

    Hydrogen isotope geochemistry of SNC meteorites

    Geochim. Cosmochim. Acta

    (1996)
  • A.S. McEwen et al.

    The rayed crater Zunil and interpretations of small impact craters on Mars

    Icarus

    (2005)
  • H.Y. McSween

    Mars

  • H.Y. McSween et al.

    Petrogenesis of the Elephant Moraine A79001 meteorite: multiple magma pulse on the shergottite parent body

    Geochim. Cosmochim. Acta

    (1983)
  • H.Y. McSween et al.

    Mars

  • R.E. Novak et al.

    Measurement of the isotopic signatures of water on Mars: implications for studying methane

    Planet. Space Sci.

    (2011)
  • L.E. Nyquist et al.

    Concordant Rb–Sr, Sm–Nd, and Ar–Ar ages for Northwest Africa 1460: a 346 Ma old basaltic shergottite related to “lherzolitic” shergottites

    Geochim. Cosmochim. Acta

    (2009)
  • A.H. Peslier et al.

    Crystallization, melt inclusion, and redox history of a Martian meteorite: olivine-phyric shergottite Larkman Nunatak 06319

    Geochim. Cosmochim. Acta

    (2010)
  • J.T. Shafer et al.

    Trace element systematics and 147Sm–143Nd and 176Lu–176Hf ages of Larkman Nunatak 06319: closed-system fractional crystallization of an enriched shergottite magma

    Geochim. Cosmochim. Acta

    (2010)
  • A.H. Treiman et al.

    The SNC meteorites are from Mars

    Planet. Space Sci.

    (2000)
  • T. Usui et al.

    Petrogenesis of olivine-phyric shergottite Yamato 980459, revisited

    Geochim. Cosmochim. Acta

    (2008)
  • T. Usui et al.

    Origin of water and mantle–crust interactions on Mars inferred from hydrogen isotopes and volatile element abundances of olivine-hosted melt inclusions of primitive shergottites

    Earth Planet. Sci. Lett.

    (2012)
  • Y.L. Yung et al.

    HDO in the Martian atmosphere: implications for the abundance of crustal water

    Icarus

    (1988)
  • C.B. Agee et al.

    Unique meteorite from early amazonian Mars: water-rich basaltic breccia Northwest Africa 7034

    Science

    (2013)
  • C.M.O.D. Alexander et al.

    The provenances of asteroids, and their contributions to the volatile inventories of the terrestrial planets

    Science

    (2012)
  • J.B. Balta et al.

    Magmatic history and parental melt composition of olivine-phyric shergottite LAR 06319: importance of magmatic degassing and olivine antecrysts in Martian magmatism

    Meteorit. Planet. Sci.

    (2013)
  • J.P. Bibring et al.

    Global mineralogical and aqueous Mars history derived from OMEGA/Mars express data

    Science

    (2006)
  • D.D. Bogard et al.

    Martian gases in an Antarctic meteorite?

    Science

    (1983)
  • Cited by (81)

    • The role of assimilation and fractional crystallization in the evolution of the Mars crust

      2022, Earth and Planetary Science Letters
      Citation Excerpt :

      % to 2 wt. % H2O (e.g., McCubbin et al., 2016; Usui et al., 2015). All H2O content estimates are given in Table 1.

    View all citing articles on Scopus
    View full text