Ancient fingerprints in the clay

The thermodynamics of ancient clays on Mars seems inconsistent with the idea that a thick atmosphere of carbon dioxide caused a warm, wet era in the planet's early history. What did cause it remains an enigma.

Volcanic rocks dominate the surface of Mars. In the cold, dry conditions that prevail there today, these rocks can persist with little chemical alteration. But the recent discovery by Mars orbiters and landers of sulphate and clay deposits on the planet indicates that Mars' ancient environment was different, and involved liquid water1. On page 60 of this issue2, Chevrier et al. add a new twist to this idea, with calculations showing how ancient martian clays formed in aqueous environments, but with little carbon dioxide present. That runs contrary to a popular view of an early Mars where liquid water was sustained by the greenhouse effect of a thick, CO2-rich atmosphere.

Evidence that Mars was wetter before about 3.7 billion years ago comes from various geomorphic features1. In particular, valleys with characteristic branching forms seem to have been eroded by water, and some evidence also argues for the influence of rainfall. The heavily degraded rims of ancient craters and crater infilling are similarly interpreted as fluvial features in some models.

But whether a warmer, wetter earlier climate on Mars was persistent, intermittent or even existed at all remains controversial. If it did exist, what caused it? The idea that Mars' early atmosphere contained much more CO2 than it does now is an appealing suggestion. More CO2 provides greater warmth through its own greenhouse effect, together with that of increased water vapour. Because the early Sun was about 25% fainter around 3.5 billion years ago than it is today, several bars of CO2 pressure would be required to achieve the necessary warming (1 bar is Earth's approximate atmospheric pressure).

At around 1.5 bar, however, CO2 ice clouds start to form under the faint Sun. It has been suggested that these clouds might themselves produce warming, but up-to-date models show that CO2 particles precipitate out or dissipate and cannot sustain a sufficient greenhouse effect3. Moreover, a substantial amount of atmospheric CO2 should have left behind carbonate deposits. But on Mars' surface, not a single carbonate outcrop has been identified down to a horizontal scale of 100 metres. One possibility is that carbonates did not form because waters were rich in sulphuric acid, which displaces CO2 (ref. 4). Although this is true in equilibrium, it still leaves the temporal evolution of CO2 in the martian atmosphere unresolved.

Chevrier et al.2 contribute to this debate by considering the thermodynamic equilibrium of clay minerals known as smectites. These clays, which have a characteristic layered-sheet structure, are found on Mars in both iron-rich and aluminium-rich forms. During the process of aqueous weathering, soluble ions are leached from 'primary' igneous rocks. 'Secondary' salts and clays, such as carbonates, sulphates and smectites, subsequently precipitate from the leaching fluid. What secondary minerals are deposited depends sensitively on the composition, pH and temperature of the leaching fluid. These minerals are consequently 'Rosetta stones' for deciphering past environments.

Chevrier et al. show that the equilibrium between carbonates and smectites implies the presence of less than 0.001 to 0.01 bar CO2 in Mars' early atmosphere for smectites to predominate as observed. Moreover, the neutral to alkaline conditions required to form the clays are inconsistent with the presence of sulphuric acid postulated to explain the absence of large carbonate deposits. According to this model, Mars' extensive sulphate deposits date from a later, more acidic geochemical era.

The lack of CO2 in the early martian atmosphere could be explained by its early atmospheric loss. Mars' gravity is two-fifths that of Earth, making two processes, hydrodynamic escape and impact erosion, effective. Hydrodynamic escape would have occurred during Mars' first few hundred million years, when atmospheric hydrogen flowed out in a 'planetary wind' — analogous to the solar wind — that dragged along and removed heavier gases. Impact erosion results from the cumulative effect of large impacting bodies releasing enough energy to blast parts of the atmosphere away. There is evidence for both processes in noble-gas isotopic ratios on Mars.

In the absence of CO2, Chevrier et al. suggest2 that other greenhouse gases promoted warmth. One candidate is methane; but methane is decomposed by ultraviolet sunlight, so a methane source comparable to Earth's biosphere would be needed to warm Mars above freezing. Another possibility is sulphur dioxide (SO2), which, judging by the bulk chemistry of martian meteorites, could have been released from martian volcanoes in amounts similar to or exceeding their water emission. But SO2 is soluble, and could have acted only as a 'lever' to raise temperatures to near freezing, making it easier for perturbations, such as asteroid or comet impacts, to cause temporary wet climates. Such impacts would have flash-heated the surface and released water, producing rain and erosion. Irrespective of any long-term greenhouse effect, the conclusion seems unavoidable that Mars was warmed transiently by many impacts early in its history5.

A pivotal assumption made by Chevrier et al. is that the clays were formed in equilibrium with the atmosphere. But if the clays were formed in isolation from the atmosphere, beneath the surface, their thermodynamics might be of little relevance. Smectites are found in martian meteorites (Fig. 1) that formed in the subsurface when small amounts of saline waters infiltrated basalt6. Generally, suitable subsurface environments include long-lived hydrothermal systems that result from impacts on ice-rich ground; indeed, on Earth, smectites are characteristic of some hydrothermal systems in impact craters7.

Figure 1: Subsurface smectites.

The yellow, reddish-brown and black veinlets in this view (about 1 mm across) of the 'Lafayette' martian meteorite are mixtures of clay and iron oxides that formed when small amounts of saline water infiltrated bedrock beneath Mars' surface and reacted with basaltic mineral grains. The presence of such clays beneath the martian surface is one caveat to interpreting Chevrier and colleagues' findings2 as clinching evidence against a carbon-dioxide-rich early martian atmosphere. (Picture courtesy of A. Treiman, Lunar and Planetary Institute.)

A further unresolved problem is that sulphates are found on Mars in places that range from young deposits around the planet's northern cap8 to the ancient Meridiani outcrops9, which are thought to be more than 3.7 billion years old. The stratigraphy of sulphate-rich deposits in an enormous chasm, Juventae Chasma, suggests that some of those deposits are similarly ancient10. The interpretation of Chevrier and colleagues' findings could therefore be more complicated than a geochemical history of an early age of clays succeeded by a sulphate era.

The history of recent Mars exploration shows that observations on the ground can completely overturn ideas inferred from orbit. Gusev Crater was chosen as a landing site for NASA's Spirit rover because it was interpreted from above as a lakebed. But what the rover found was a largely basaltic surface, rather than fluvial sediments. Meridiani Planum was selected for NASA's Opportunity rover because orbital data showed abundant haematite. But surface minerals there proved to be even richer in sedimentary sulphates.

Only a complete picture on the ground can provide confidence about the early environment of Mars, and whether it was ever conducive to life. Two rovers, NASA's Mars Science Laboratory rover and the European Space Agency's ExoMars, planned for launch in 2009 and 2013, respectively, might supply the answers — if targeted to the clays and sulphates of the planet.


  1. 1

    Carr, M. H. The Surface of Mars (Cambridge Univ. Press, 2006).

  2. 2

    Chevrier, V., Poulet, F. & Bibring, J.-P. Nature 448, 60–63 (2007).

  3. 3

    Colaprete, A. & Toon, O. B. J. Geophys. Res. Planets 108, doi:10.1029/2002JE001967 (2003).

  4. 4

    Fairen, A. G., Fernandez-Remolar, D., Dohm, J. M., Baker, V. R. & Amils, R. Nature 431, 423–426 (2004).

  5. 5

    Segura, T. L., Toon, O. B., Colaprete, A. & Zahnle, K. Science 298, 1977–1980 (2002).

  6. 6

    Treiman, A. H. & Lindstrom, D. J. J. Geophys. Res. Planets 102, 9153–9163 (1997).

  7. 7

    Naumov, M. V. Geofluids 5, 165–184 (2005).

  8. 8

    Langevin, Y., Poulet, F., Bibring, J.-P. & Gondet, B. Science 307, 1584–1586 (2005).

  9. 9

    Squyres, S. W. et al. Science 313, 1403–1407 (2006).

  10. 10

    Catling, D. C. et al. Icarus 181, 26–51 (2006).

Download references

Author information

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Catling, D. Ancient fingerprints in the clay. Nature 448, 31–32 (2007).

Download citation

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.