Martian water stored underground

Why did Mars lose so much of its surface water, whereas Earth retained its? Models of the evolution of minerals on the two planets suggest one explanation: the Martian water was drawn into the planetary interior.

Observations of Mars and its geology suggest that the red planet once had an Earth-like hydrological cycle that included large lakes or oceans. In contrast to this ancient wet environment, the surface of Mars today is cold and dry. The transition to this present state is closely linked to the fate of the planet’s surface water, which is poorly understood. A substantial amount of surface water escaped to space from the atmosphere, in part because of Mars’ relatively low gravity1. However, atmospheric-escape models account for only some of ancient Mars’ surface water. On page 391, Wade et al.2 propose that much of the surface water was sequestered underground.

Erupted lava had the chance to interact with surface water to form hydrated crust on both Earth and ancient watery Mars. Wade et al. examine the thermodynamic properties of the water-bearing ‘mafic’ crusts (which largely consist of the rock basalt) of each planet, and show that Martian basalt can hold more water than terrestrial basalt, and can effectively transport it to a greater depth below the surface (more than 90 kilometres; Fig. 1). The authors also compute the stability of water-containing minerals in hydrated crusts and their bulk-rock densities along both planets’ geotherms (which describe how temperature varies with depth). They conclude that the burial of hydrated crusts progressively hydrates the interior of Mars, but that this process does not work effectively for Earth.

Figure 1 | The water content of basalt on Earth and Mars. Wade et al.2 report modelling of the surface evolution of Earth and Mars, which allowed them to estimate the amount of water that can be sequestered by basalt rock in the planets’ interiors. They find that the Martian basalts can store more water (shown as percentage by weight), and at greater depths, than their terrestrial counterparts. This suggests that much of the water thought to have existed on the surface of ancient Mars was sequestered underground. (Adapted from Fig. 1a of ref. 2.)

The four rocky planets of the Solar System (Mercury, Venus, Earth and Mars) are thought to have been formed by the accretion of similar planetary building blocks. This resulted in their mantles having broadly similar compositions of all the major elements, except for iron. Metallic iron partitions into the metallic cores of the rocky planets, whereas iron(ii) oxide accumulates in silicate-rich planetary mantles. A thermodynamic property known as oxygen fugacity, which is a measure of the amount of oxygen present in a mixture, controls iron content in planetary basalts. Martian basalts have a distinctly higher content of iron(ii) oxide (about 17% by weight) than do typical basalts on Earth (about 7–10% by weight), which suggests that oxygen fugacity was higher during Martian core formation than it was during Earth’s core formation. Wade and colleagues show that this compositional difference, along with the different geotherms of Mars and Earth, has a key role in the storage and transportation of planetary surface water in the crust and mantle of the two planets.

Hydration processes generally expand the crustal volume, making the crust less dense. By contrast, Wade and colleagues’ thermodynamic modelling indicates that the iron-rich Martian basalts undergo small volume expansions during hydration. Furthermore, hydrated iron-rich Martian basalts tend to melt at lower temperatures (about 800–900 °C) than does anhydrous mafic rock (about 1,200 °C), and this melting leaves relatively dense hydrated residues in the mantle. In the apparent absence on Mars of tectonic processes that recycle crust material into the mantle, the authors propose that successive burial of hydrated crust might have induced hydration of the mantle — such burial would have gradually increased the temperature and pressure applied to hydrated crustal basalts, causing them to melt, and therefore to leave hydrated residues in the mantle.

Hydration of the Martian mantle would also lead to it becoming more oxidized. However, geochemical analysis of meteorites (known as shergottites) formed from young Martian basalts suggest that their source in the mantle is less oxidized than Earth’s mantle3. Moreover, the shergottite source is depleted in water (less than 50 parts per million)4 relative to Earth’s mantle (typically about 100–200 p.p.m.)5. How do these observations fit into Wade and colleagues’ suggestion that the Martian mantle contains more water than Earth’s? Experiments have shown6 that the shergottite source is located at a depth of approximately 100 km, which is near the base of the hydrated mantle column proposed by Wade and co-workers. The lower oxidation and hydration of the shergottite source might therefore be representative of (or place a cap on the state of) the uppermost region of the unhydrated mantle, unless it represents a local phenomenon.

The global surface-water inventory of Mars was originally estimated to be about 2 × 107 to 2 × 108 km3 (ref. 7, and references therein), on the basis of the size of the putative ancient oceans. But estimates of the total water loss to space are much smaller (less than 107 km3, based on atmospheric-escape models7). The discrepancy between these estimates hints at the existence of a ‘missing’ water reservoir beneath the surface. The widespread distribution of hydrated materials on the surface of Mars also implies the existence of a crustal water reservoir, but conventional spectroscopic observations are able to see only the surface veneer. Wade and colleagues’ thermodynamic modelling, together with remote-sensing observations, offers a means to work out the depth profile of hydrated materials, and to calculate a reasonable estimate of the volume of the crustal water reservoir.

Ground ice might also account for the missing water reservoir on Mars810. Subsurface radar-sounder measurements8 have detected an anomaly in an electrical property of rocks in the planet’s northern hemisphere, which implies that massive ice deposits are embedded among or between layers of sediment and volcanic materials at a depth of 60–80 m. The ground-ice model has also been proposed on the basis of analyses of hydrogen isotopes in Martian meteorites9 and of crater morphology10. The crater study indicates that the subsurface water ice has a volume of about 3 × 107 km3, which is comparable to the size of the ancient oceans. Subsurface exploration will be required to test both the hydrated-crust and ground-ice theories, and therefore to shed light on the evolution of the Martian water inventory.

doi: 10.1038/d41586-017-08670-y
Nature Briefing

Sign up for the daily Nature Briefing email newsletter

Stay up to date with what matters in science and why, handpicked from Nature and other publications worldwide.

Sign Up


  1. 1.

    Lammer, H. et al. Space Sci. Rev. 174, 113–154 (2013).

  2. 2.

    Wade, J., Dyck, B., Palin, R. M., Moore, J. D. P. & Smye, A. J. Nature 552, 391–394 (2017).

  3. 3.

    Herd, C. D. K. et al. Geochim. Cosmochim. Acta 66, 2025–2036 (2002).

  4. 4.

    Usui, T., Alexander, C. M. O’D., Wang, J., Simon, J. I. & Jones, J. H. Earth Planet. Sci. Lett. 357–358, 119–129 (2012).

  5. 5.

    Saal, A. E., Hauri, E. H., Langmuir, C. H. & Perfit, M. R. Nature 419, 451–455 (2002).

  6. 6.

    Filiberto, J. & Dasgupta, R. J. Geophys. Res. Planets 120, 109–122 (2015).

  7. 7.

    Kurokawa, H. et al. Earth Planet. Sci. Lett. 394, 179–185 (2014).

  8. 8.

    Mouginot, J., Pommerol, A., Beck, P., Kofman, W. & Clifford, S. M. Geophys. Res. Lett. 39, L02202 (2012).

  9. 9.

    Usui, T., Alexander, C. M. O’D., Wang, J., Simon, J. I. & Jones, J. H. Earth Planet. Sci. Lett. 410, 140–151 (2015).

  10. 10.

    Weiss, D. K. & Head, J. W. Icarus 288, 120–147 (2017).

Download references