News & Views | Published:

Planetary science

Water cycling on Mars

Nature volume 446, pages 150151 (08 March 2007) | Download Citation


The Meridiani Planum region on Mars is rich in minerals derived from evaporation, but lacks a topography consistent with standing water. Do the deposits stem from upwelling groundwater early in the planet's history?

A succession of sophisticated spacecraft missions has led to spectacular advances in the understanding of Mars' global hydrology over the past few decades. One of many examples is the discovery of abundant hydrated sulphate salt minerals. These minerals are found at many locations on the planet — most notably at Meridiani Planum, the landing site of NASA's robotic Mars rover Opportunity — and prove that water must once have been abundant on the surface of Mars. On page 163 of this issue, Andrews-Hanna et al.1 use a numerical model to simulate the evolving global flow of subsurface groundwater early in Mars' geological history. They place their simulation in the context of the formation of the enormous volcanic uplift feature known as Tharsis.

One way of developing a model of martian hydrology comes from a comparison with what we know about Earth. Western science was painfully slow in achieving its understanding of Earth's hydrological cycle. Many, if not most, of Isaac Newton's scientific contemporaries held the view that Earth's rivers were ultimately fed from upland springs. The springs were presumed to discharge water from within the planet, in much the same way as blood flows from a cut in an artery of the human body. Water from the oceans was presumed to return to the land through subsurface veins.

By contrast, Eastern philosophical writings had long held that Earth's water flowed as part of a great cycle involving the atmosphere. About 3,000 years ago, the Vedas, Hindu texts of ancient India, explained Earth's water movements in terms of cyclical processes of evaporation, condensation, cloud formation, rainfall, river flow and water storage2. This concept of a global water cycle entered Western thought only late in the seventeenth century, when Edmond Halley, among others, showed that the evaporation from Earth's oceans supplied the rain clouds that led to a balancing run-off of water from land to the seas. Specifically, Halley compared evaporation from the Mediterranean Sea with estimates of its river inflow, thereby providing a modern scientific expression of the hydrological cycle.

Earth is a relatively warm and wet planet with a substantial atmosphere. So how does its water cycle relate to that of Mars, with that planet's extremely cold and dry climate and tenuous atmosphere? Although the conditions on Mars' present-day surface do not favour the persistence of exposed surface water or ice, high-resolution imagery of the planet has revealed distinctive, very ancient landforms typical of erosion by water and ice occurring at the planet's surface3. The evidence from surface features, such as fluvial valley networks and giant flood channels, strongly suggests that, early in its geological history, Mars was a 'water planet' like Earth4. Nevertheless, this hypothesis was not widely accepted until 2002, when NASA's Mars Odyssey mission revealed abundant subsurface ice on the planet5.

Since then, products of the alteration of clay minerals by water6 and evaporite salts7 — minerals left behind when water evaporates — have been identified geochemically, further confirming that the early history of Mars involved liquid water. Most spectacularly, the Burns formation, a 7-metre-thick exposure analysed by the rover Opportunity, provides a clear geological record of evaporation in an arid surface environment8 (Fig. 1). The chemistry and mineralogy of the deposit imply a fluctuating water table with acidic groundwater9. Thus, shallow, temporary, evaporating pools would have formed on the planet's surface, in much the same manner as they form under similar conditions on Earth at White Sands in the New Mexico desert8. But the Meridiani Planum region, where Opportunity detected mineral deposits from evaporation, is a tilted plain lacking topographic basins in which large bodies of water could stand and evaporate. How, then, did water get to this region of Mars?

Figure 1: A well run dry?
Figure 1

The Burns formation of Meridiani Planum is rich in evaporites. It is one of the sites where groundwater came to the surface and evaporated early in Mars' history, according to a model of the planet's hydrography produced by Andrews-Hanna et al.1. Earlier evidence from the Mars rover Opportunity for past evaporation of liquid water at the site had been difficult to explain in the absence of a topographic basin where water could stand. In the right foreground lies the Wopmay rock, whose distinctive lumpy appearance could be due to cracking processes caused by exposure to water.

Andrews-Hanna et al.1 find that the Burns formation of Meridiani Planum is in fact exactly where a global groundwater simulation model for early Mars predicts upwelling of water on the ancient planet's surface. The picture is not dissimilar to the early ideas about Earth's subsurface veins of water that preceded Edmond Halley's investigations. Indeed, pictures from the new High Resolution Imaging Experiment on NASA's Mars Reconnaissance Orbiter show rocks that were once below the surface and have been altered through the influence of flowing water10. This evidence of subsurface water flow was found along fractures in Mars' low-lying sedimentary deposits, thereby confirming an ancient groundwater circulation.

Andrews-Hanna and colleagues suggest that, in its early history, Mars had a globally connected groundwater system. The water would have circulated for an extended period of time, leaching solutes from the subsurface rocks. In this model, the water table reaches the surface at locations where the surface slopes down towards the northern lowlands of Mars. This occurs, for example, in the Meridiani Planum region, which lies just south of Mars' equator. In this way, groundwater would have been supplied continuously from the subsurface, its chemical evolution reflected in the dissolved salts. When the water evaporated, the salt would have been deposited at the planet's surface. The process occurred on an immense scale, deriving subsurface flow from large areas of Mars' highlands.

Critical to the long-term evolution of subsurface water flow on Mars was the development of the Tharsis rise. Early in martian geological history, this huge volcanic bulge developed on the planet's surface through immense outpourings of basaltic lava that also injected prodigious amounts of water and carbon dioxide into the young planet's atmosphere11. The huge pile of volcanic rocks at Tharsis progressively warped the surface of the planet and, as the authors show1, thereby altered the groundwater flow. Precipitation from the very early Earth-like atmosphere recharged groundwater in the higher regions. The water then moved through underground aquifers to discharge points such as Meridiani Planum.

The simulations by Andrews-Hanna et al.1 solve the mystery of evaporite minerals in the absence of a topographic basin that would facilitate evaporation, and they help us to understand the role of topographic changes in Mars' ancient water flow. A final note of interest is that emerging subsurface acidic waters on Earth, such as those at Rio Tinto in southwestern Spain, are host to diverse living microorganisms12. Could Mars similarly have had a subsurface biosphere of microorganisms, which might have retreated to their groundwater refuge when water failed to emerge on the planet's surface during its later geological history?


  1. 1.

    , & Nature 446, 163–166 (2007).

  2. 2.

    Hydrology in Ancient India (Natl Inst. Hydrol., Roorkee, India, 1990).

  3. 3.

    Nature 412, 228–236 (2001).

  4. 4.

    The Channels of Mars (Univ. Texas Press, Austin, 1982).

  5. 5.

    et al. Science 296, 81–85 (2002).

  6. 6.

    et al. Science 312, 400–404 (2006).

  7. 7.

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

  8. 8.

    et al. Earth Planet. Sci. Lett. 240, 11–72 (2005).

  9. 9.

    et al. Earth Planet. Sci. Lett. 240, 95–121 (2005).

  10. 10.

    & Science 315, 983–985 (2007).

  11. 11.

    & Nature 412, 237–244 (2001).

  12. 12.

    et al. Earth Planet. Sci. Lett. 240, 149–167 (2005).

Download references

Author information


  1. Victor R. Baker is in the Department of Planetary Sciences and the Department of Hydrology and Water Resources, University of Arizona, Tucson, Arizona 85721, USA.

    • Victor R. Baker


  1. Search for Victor R. Baker in:

About this article

Publication history



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.

Newsletter Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing