Oceans on Mars formed early

The geometry of putative ancient shorelines on Mars suggests that these features were deformed by the growth of a massive volcanic region — a finding that has implications for the climate, geology and hydrology of early Mars.
Maria T. Zuber is in the Department of Earth, Atmospheric and Planetary Sciences, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139, USA.

Search for this author in:

Arguably the most compelling evidence that Mars once had oceans is the presence of potential shorelines in the planet’s northern hemisphere1. However, these features do not have a constant elevation, which would argue against an oceanic origin2. In a paper in Nature, Citron et al.3 show that the deviations can be explained if the shorelines formed before and during the growth of Tharsis — a volcanic region that spans nearly one-quarter of Mars’s surface area. The results imply that the formation of the planet’s oceans occurred earlier than previously thought, and might be linked to the volcanism of Tharsis and the formation of valley networks4.

The history of liquid water on Mars is of great interest because water would have strongly influenced the planet’s climate and its potential for habitability. Evidence for a watery past in Mars’s geological record is abundant, but many of the data are indicative of catastrophic outflows or intermittent releases that neither require a temperate surface environment, nor are thought to have persisted long enough for the development of microbial life5.

Researchers have long debated the possibility that a liquid-water ocean or oceans once filled the topographic depression in Mars’s northern hemisphere6. Comparisons of the roughness7 and intrinsic electrical properties (dielectric properties)8 of the proposed Martian sea floor with those of similarly sediment-laden surfaces on Earth, as well as the concentrations of certain isotopes on Mars9, support the idea of past oceans. But perhaps the most convincing evidence comes from geomorphological observations of potential ancient shorelines. In particular, geological mapping of spacecraft images has revealed putative shorelines that can be tracked for thousands of kilometres on Mars’s northern plains1.

The identification of these shorelines is not without caveats. For instance, the geomorphological observations are open to interpretation2. Furthermore, shorelines, like beaches on Earth, should trace surfaces of constant gravitational potential (such as sea levels). However, Mars’s putative shorelines show changes in elevation of up to several kilometres, indicating that if these features indeed formed at ocean coasts, they subsequently underwent substantial deformation. If the event that produced such deformation could be identified, it would constrain the relative timing of the formation of the shorelines, and therefore of the oceans.

Citron and colleagues use shoreline modelling to show that the observed deformation can be explained by the growth of Tharsis, most of which occurred some 3.6 billion or more years ago, during an early stage in Mars’s history called the Noachian period. The authors considered two extensive shorelines: Arabia and Deuteronilus (Fig. 1). Their analysis indicates that the Arabia shoreline formed before or in the early stages of Tharsis’s growth, whereas the Deuteronilus shoreline arose during the late stages. These findings are in contrast to general thinking that the oceans, if they existed, formed after Tharsis’s emplacement. By correcting for deformation events, Citron et al. place lower limits on the ocean volumes associated with the Arabia and Deuteronilus shorelines, of about 11 and 3 times that of Earth’s Mediterranean Sea, respectively.

Figure 1 | Mars’s topography and putative ancient shorelines. The surface of Mars contains a massive region called Tharsis (red), which was formed by volcanic activity, and lowland areas in the northern hemisphere (blue) that have been proposed to be the location of ancient oceans5. Geological mapping1 has revealed two extensive potential past shorelines: Arabia (dashed line) and Deuteronilus (solid lines). Citron et al.3 report that the geometry of these shorelines can be explained if they were deformed in response to the growth of Tharsis, rather than after Tharsis’s formation as previously assumed. The colour scale denotes elevation in kilometres. (Adapted from Extended Data Fig. 2 of the paper3.)

A previous study10 showed that the elevation patterns of the Arabia and Deuteronilus shorelines could be explained by the deformation induced by a reorientation of Mars’s spin axis — a phenomenon known as true polar wander11. Rotational dynamics dictates that a planet spins stably about its axis of minimum inertia. Consequently, the load associated with an ocean of the calculated volume could have resulted in true polar wander if Tharsis had formed far from Mars’s equator. However, evidence now indicates that Tharsis probably originated near the equator12,13, in which case, ocean loading would not have been great enough to destabilize the planet’s spin axis. If Tharsis had formed close to, but not precisely at, the equator, loading would have induced a limited amount of true polar wander, but not enough to explain the deformation of the proposed shorelines.

Citron and colleagues’ scenario is therefore more consistent with our current understanding of Tharsis’s emplacement. The authors’ revised sequence of events also has some appeal, in that it ties ocean formation to the release of water-rich gas from massive, extended volcanic events. And, in the case of the ocean associated with the Arabia shoreline, it links ocean formation to the emergence of valley networks4 — abundant features on early Mars that carried water to the northern lowlands in which the oceans were situated.

Not surprisingly, the interplay of water and geodynamics over Mars’s history is more complicated than presented in Citron and colleagues’ analysis. During the period of interest, additional deformational processes occurred at local to global length scales that were not accounted for. Such processes are associated with, for example, mantle activity, impacts of asteroids or comets on to Mars’s surface, glaciers, erosion and individual volcano growth, and could also have contributed to the deformation of the shorelines.

More-detailed mapping and modelling could clarify the contributions to the observed elevation changes and refine the sequence and relative timing of events. In addition, other, much shorter and more fragmented potential shorelines have been proposed14,15 that suggest the existence of seas and lakes, and that merit further study. Finally, the timing and role of volcanism, as it relates to the state and evolution of the atmosphere and to the persistence of water on the planetary surface, is far from understood. But, fortunately, the extensive database of remote and in situ observations of Mars continues to accumulate, revealing the intriguing story of Mars’s water-rich past.

Nature 555, 590-591 (2018)

doi: 10.1038/d41586-018-03415-x
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.

    Parker, T. J., Gorsline, D. S., Saunders, R. S., Pieri, D. C. & Schneeberger, D. M. J. Geophys. Res. Planets 98, 11061–11078 (1993).

  2. 2.

    Malin, M. C. & Edgett, K. S. Geophys. Res. Lett. 26, 3049–3052 (1999).

  3. 3.

    Citron, R. I., Manga, M. & Hemingway, D. J. Nature 555, 643–646 (2018).

  4. 4.

    Bouley, S. et al. Nature 531, 344–347 (2016).

  5. 5.

    Carr, M. H. Water on Mars 197 (Oxford Univ. Press, 1996).

  6. 6.

    Baker, V. R. et al. Nature 352, 589–594 (1991).

  7. 7.

    Head, J. W. III et al. Science 286, 2134–2137 (1999).

  8. 8.

    Mouginot, J., Pommerol, A., Beck, P., Kofman, W. & Clifford, S. M. Geophys. Res. Lett. 39, l02202 (2012).

  9. 9.

    Villanueva, G. L. et al. Science 348, 218–221 (2015).

  10. 10.

    Perron, J. T., Mitrovica, J. X., Manga, M., Matsuyama, I. & Richards, M. A. Nature 447, 840–843 (2007).

  11. 11.

    Goldreich, P. & Toomre, A. J. Geophys. Res. 74, 2555–2567 (1969).

  12. 12.

    Roberts, J. H. & Zhong, S. Icarus 190, 24–31 (2007).

  13. 13.

    Matsuyama, I. & Manga, M. J. Geophys. Res. Planets 115, E12020 (2010).

  14. 14.

    Clifford, S. M. & Parker, T. J. Icarus 154, 40–79 (2001).

  15. 15.

    Carr, M. H. & Head, J. W. III J. Geophys. Res. Planets 108, 5042 (2003).

Download references