Mars Water: Evidence That the Red Planet Was Once Covered in Oceans
Valleys, clay minerals and lake-bed rocks show Mars was once wet. Whether it held a northern ocean is unsettled, and so is where the water went. A review of the evidence.
Present-day Mars is a cold, thin-aired desert. Surface pressure is under one percent of Earth's, and liquid water is not stable in the open. Yet the planet's oldest terrains record a very different climate. The live research question is no longer whether liquid water once shaped the surface, but how much of it there was, how continuously it persisted, and where it went. "Mars had oceans" is the most extreme answer on offer, and also the least settled. This piece separates what orbital and rover data establish from what is still interpretation.
What the rocks and minerals establish
Three independent lines of evidence point to a wetter past. The first is morphology: branching valley networks, mostly in ancient, heavily cratered terrain dated to the Noachian and early Hesperian periods (roughly more than 3.5 billion years ago). Their drainage patterns resemble those cut by runoff and groundwater sapping, although how long the flows lasted is debated.
The second is mineralogy. Infrared spectrometers on Mars Express (OMEGA) and Mars Reconnaissance Orbiter (CRISM) found widespread phyllosilicates, or clay minerals, in the oldest crust, and sulfates in somewhat younger layered deposits (Bibring and colleagues, Science, 2006). Clays generally form when rock reacts with water at near-neutral conditions; sulfates point to later, more acidic and evaporative settings. The sequence suggests a planet that was wetter early and became drier and more acidic over time, though it is a global pattern inferred from patchy exposures.
The third is direct sedimentology from the ground. The Opportunity rover found cross-bedded sandstones and chemical signatures at Meridiani Planum consistent with sediment that was reworked by water and later altered by it (Squyres and colleagues, Science, 2004). Curiosity, in Gale crater, drilled fine-grained mudstones at Yellowknife Bay that were interpreted as lake-bed deposits from an environment with the chemical ingredients and mild conditions that microbes on Earth could use (Grotzinger and colleagues, Science, 2014). Perseverance is exploring the delta of an ancient river that fed Jezero crater. A delta is a particularly useful target because it concentrates fine sediment, which is the material most likely to preserve organic compounds.
The ocean hypothesis and why it remains contested
The idea that a large ocean once filled the northern lowlands rests mainly on the unusually smooth, low-lying terrain there and on proposed shorelines traced around its edge. Di Achille and Hynek (Nature Geoscience, 2010) argued that river deltas cluster near a common elevation, as they would if they had formed at a single sea level. Perron and colleagues (Nature, 2007) proposed that the shorelines are tilted because Mars's rotation axis has shifted and the Tharsis volcanic province has deformed the crust since then, so that an originally level shoreline would no longer look level.
The counterarguments are substantial. High-resolution images did not show clear, continuous beach-like features along the proposed coastlines (Malin and Edgett, Geophysical Research Letters, 1999), and a shoreline that has to be corrected for large later deformation is hard to test. An ocean also needs a climate that allows liquid water over long periods, which brings the faint young Sun problem into play: the early Sun is expected to have been dimmer, so something other than sunlight, perhaps a thick carbon dioxide atmosphere with additional greenhouse gases such as hydrogen, must have kept Mars above freezing at least episodically (see Ramirez and colleagues, Nature Geoscience, 2014; Wordsworth, Annual Review of Earth and Planetary Sciences, 2016). Many researchers favour a "cold and icy" early Mars with episodic melting, rather than a persistently warm one. At present the evidence supports lakes, rivers, and groundwater with confidence; a planet-scale ocean is a hypothesis.
Where did the water go?
Three sinks are usually considered, and they are not mutually exclusive.
Escape to space
NASA's MAVEN orbiter measures how the upper atmosphere interacts with the solar wind and how fast gas leaves. Its observations of argon isotopes indicate that a large fraction of the original atmosphere has been lost to space (Jakosky and colleagues, Science, 2017), and the measured present-day loss rates, integrated back in time with models, indicate that escape was a major process (Jakosky and colleagues, Icarus, 2018). Hydrogen escape also leaves a fingerprint: the ratio of deuterium to hydrogen in Martian water is several times the terrestrial value, because the lighter isotope escapes more easily.
The magnetic field
Spacecraft magnetometers found strongly magnetised patches of ancient crust, which implies Mars once had an internal dynamo that later shut down (Acuna and colleagues, Science, 1999). The popular summary, that losing the field "stripped" the atmosphere, is too simple. A global field is not the only control on escape, and how much the missing shield mattered compared with, for example, the weak gravity of a small planet and intense early solar activity remains under study.
Locked in the crust and frozen
Scheller and colleagues (Science, 2021) estimated that hydrated minerals in the crust could have sequestered a large part of the planet's early water, with the rest lost to space. Their work is a model-based argument, but it shifts attention from the sky to the ground: if clays and similar minerals hold the water, then some of it is still there, only not as a liquid on the surface. Ice is the other reservoir, in the polar caps and in mid-latitude deposits detected by orbital radar and confirmed in situ near the surface by the Phoenix lander in 2008.
Water today, and the subsurface liquid question
Surface liquid water is not stable under current pressure and temperature, though thin films and brines could exist briefly. Radar sounding by MARSIS reported a bright reflector under the south polar layered deposits in 2018 that was interpreted as a body of liquid water (Orosei and colleagues, Science, 2018). That reading is disputed: other researchers have argued that clays, salts, or other materials could produce similar radar signals without liquid. A separate analysis of seismic data from the InSight lander proposed that the mid-crust could contain liquid water in pore spaces (Wright and colleagues, PNAS, 2024). It is an interpretation that depends on rock-physics assumptions. Neither claim has been independently confirmed by drilling or a second type of measurement.
Limitations and open questions
- Duration versus abundance. A river valley records that water flowed, not for how many years. Estimates of how long Jezero's delta took to form vary with assumed discharge and sediment supply.
- Habitable is not inhabited. Water, energy, and the right chemistry make an environment habitable. No mission has found evidence that life arose on Mars.
- Dating. Age estimates from crater counting have large uncertainties. Radiometric ages from returned samples would calibrate the whole chronology.
- Preservation. Organic molecules are destroyed by radiation and oxidants near the surface, so a negative detection is weak evidence, and a positive detection could have a non-biological origin.
What researchers should watch
- The analysis of the rock cores Perseverance has already cached. The planned return of samples to Earth would allow laboratory isotope work that no rover can do, although the mission's schedule and funding have been under review.
- Whether the deuterium-to-hydrogen ratio in ancient clays differs from that in the modern atmosphere. This would constrain how much water escaped and when.
- Independent checks of subsurface liquid claims, for example with a different radar frequency or seismic reanalysis.
- Better coupling of climate models with geology: a model that makes rivers should also explain why they stopped.
For now the defensible summary is that early Mars was episodically wet enough to form lakes, deltas, and groundwater systems, and that the case for a standing northern ocean remains open. Each of these points can be tested with data, which is why the planet remains such a productive target.
References and further reading
- Bibring, J.-P. et al. Global mineralogical and aqueous Mars history derived from OMEGA/Mars Express data. Science, 2006.
- Squyres, S. W. et al. In situ evidence for an ancient aqueous environment at Meridiani Planum, Mars. Science, 2004.
- Grotzinger, J. P. et al. A habitable fluvio-lacustrine environment at Yellowknife Bay, Gale crater, Mars. Science, 2014.
- Di Achille, G. and Hynek, B. M. Ancient ocean on Mars supported by global distribution of deltas and valleys. Nature Geoscience, 2010.
- Perron, J. T. et al. Evidence for an ancient martian ocean in the topography of deformed shorelines. Nature, 2007.
- Malin, M. C. and Edgett, K. S. Oceans or seas in the Martian northern lowlands: high resolution imaging tests of proposed coastlines. Geophysical Research Letters, 1999.
- Jakosky, B. M. et al. Mars' atmospheric history derived from upper-atmosphere measurements of 38Ar/36Ar. Science, 2017.
- Scheller, E. L. et al. Long-term drying of Mars by sequestration of ocean-scale volumes of water in the crust. Science, 2021.
- Wordsworth, R. D. The climate of early Mars. Annual Review of Earth and Planetary Sciences, 2016.