water on Mars

new gully deposit on Mars

Two images taken by Mars Global Surveyor in Aug. 1999 (left) and Sep. 2005 (right) of gullies on the wall of a crater in the Centauri Montes region of Mars. The later one shows the appearance of new, light-colored deposit, possibly due to running water.

Ravi Vallis

Viking mosaic of the Ravi Vallis channel. The region shown is about 360 km (225 miles) long. Like many other channels that empty into the northern plains of Mars, Ravi Vallis originates in a region of chaotic (collapsed and disrupted) terrain within the planet's older, cratered highlands. Structures in these channels indicate that they were carved by liquid water moving at high flow rates – up to 1000's of times the outflow of the Amazon river. The abrupt beginning of the channel, with no apparent tributaries, suggests that the water was released under great pressure from beneath a confining layer of frozen ground. As this water was released and flowed away, the overlying surface collapsed, producing the disruption and subsidence shown here. Three such regions of chaotic collapsed material are seen in this image, connected by a channel whose floor was scoured by the flowing water. The flow in this channel was from west to east (left to right). This channel ultimately links up with a system of channels that flowed northward into Chryse Basin. Image & caption: LPI.

How much liquid water existed on Mars in the past, and when? Does liquid water exist on Mars today? These are questions that fascinate scientists, especially astrobiologists, because they have a direct bearing on whether there was once life on Mars and, if so, whether it has survived to the present day.


Images taken by Mars Global Surveyor (MGS) have provided some of the best evidence yet that water still occasionally flows on the Martian surface. Two gullies on the inside of craters, which were originally photographed by MGS in 1999 and 2001, and imaged again in 2004 and 2005, showed changes consistent with water flowing down the crater walls, according to a study published in December 2006.1 The two sites are in the Terra Sirenum and the Centauri Montes regions of southern Mars. In both cases, researchers found bright, light-colored deposits, each several hundred meters long, in the gullies that were not present in the original photos. They concluded that the deposits – possibly mud, salt, or frost – were left there when water recently cascaded through the channels.


gully flow in Terra Sirenum crater
Enlarged view of a new gully deposit in a crater in Terra Sirenum.


Other scientists challenged this explanation, arguing that the atmosphere of Mars is so thin and the temperature so cold that liquid water couldn't persist at the surface but would rapidly evaporate or freeze. Some of these scientists suggested that the gullies, and others like them discovered by MGS, might have been caused not by water but by liquid carbon dioxide, which has a lower melting point than water. However, the carbon dioxide theory does not bear close scrutiny, for the following reason. The minimum pressure under which a substance can exist as a liquid occurs at the substance's triple point. In the case of carbon dioxide, the triple point corresponds to a temperature of -56.6°C and a pressure of 5.1 atmospheres (518,000 Pa). Now, while Mars does get as cold as (or colder than) -56.6°C, the surface pressure on Mars (or on Earth, for that matter) certainly never reaches 5.1 atmospheres. Thus we can be absolutely certain that carbon dioxide cannot exist as a liquid on or anywhere near the surface on Mars. In contrast, the triple point of water occurs at a mere 611.73 Pa (6.1173 millibars), a pressure that is frequently exceeded on the surface of Mars, in craters and other low-lying areas, according to measurements by Viking and other spacecraft. This fact adds credence to the view that water could remain liquid long enough, after breaking out from an underground source, to carry debris downslope before freezing.


Water in the past on Mars

In previous centuries, astronomers thought that the dark areas they saw on Mars through their telescopes might be seas, and, at the end of the nineteenth century there was much speculation, led by Percival Lowell, about the possibility of Martian canals. By the the early decades of this century, however, it had become clear, because of the very low atmospheric pressure, that little or no surface water could exist on the Red Planet today.


In 1969, Mariner 9 provided the first strong evidence that liquid water had flowed on the surface of Mars in the remote past. Among the thousands of images it sent back from orbit were those of flat-floored channels with eroded banks, sand bars and teardrop-shaped islands, channels with second- and third-order tributary systems, and braided channels which, had they been encountered on Earth, would unhesitatingly have been attributed to episodic flooding. Later probes have added to the evidence of water-carved channels and other features on Mars.


The fact that Mars had flowing water implies that conditions on the planet were once very different than they are today (see Mars, past conditions). Yet, curiously, there are no signs that it ever rained on the fourth planet. The water-carved systems on Mars are short and stubby, dividing little upstream, and ending abruptly as if the water had suddenly appeared at that spot rather than having fallen over a large area and become collected. The assumption is that the Martian water erupted from beneath the surface, welling up as a result of volcanic eruptions or asteroid impacts, and then flooded to form channels, and lakes, and perhaps even seas.


An ocean on Mars?

If there were large bodies of water long ago on Mars, where were they, how big were they, and how long did they last? According to one theory, a great ocean once filled the low, flat plains of the northern hemisphere. Today these gently sloping plains are marked by ridges, low hills, and sparsely scattered craters. They are considered to be volcanic in origin and Hesperian (3.5 to 1.8 billions years) in age based on crater counts, but they are much smoother than comparable plains in the south and covered by layers of younger material at least 100 meters thick.


hypothetical ocean on Mars
What Mars would look like today if it still had the huge amount of water some researchers believe was present at its formation. Image: NASA


In 1989, Timothy Parker of the Jet Propulsion Laboratory and colleagues argued that there was geological evidence of a shorelines.2 The originally proposed shorelines were two discontinuous boundaries between landforms thought to have formed by wave or other water-related processes. Stephen Clifford (Lunar and Planetary Institute) and Parker later refined the outlines and hypothesized that Noachian-aged (3.8 to 3.5 billion year old) bodies of water and ice covered up to one third of the surface of Mars.3 The two most continuous supposed shorelines, called the Arabia and Deuteronilus shorelines, run roughly parallel with the southern boundary of the northern plains. The Arabia shoreline can be traced all around the planet except through the Tharsis region. But its elevation varies by several kilometers, in some places by 11 km – a large range that doesn't fit well with a shoreline interpretation. Features of the proposed shoreline that have been interpreted as formed by wave actions or other marine processes can be equally argued as being formed by mass wasting and volcanic processes. The Deuteronilus boundary is less conspicuous than the Arabia one but has a smaller range in elevations. For nearly half its length the Deuteronilus boundary marks the southern extent of the geologic unit called the Vastitas Borealis Formation. For the rest of its length it is seen only intermittently around clusters of hills or across lava flows. The trouble is, if Deuteronilus is an ancient shoreline, why does it lack key features, such as inward-facing cliffs or channels that end abruptly as they enter a large body of water, that would be expected at the margins of a great sea? The fact is, the shoreline evidence for a northern ocean is not all that convincing.


Evidence from the rovers

spherules in rock on Meridiani Planum
Spherules found in rock on Meridiani Planum - probably concretions formed as water evaporated


In March 2004, NASA scientists involved with the Mars Exploration Rovers put forward evidence that rocks near the Opportunity landing site, on Meridianii Planum, had once been drenched in water – the most convincing and powerful case yet made for long-standing liquid water on the surface of Mars. Several key lines of evidence support this conclusion, including the presence in the rocks of sulfates and small spherules that were probably precipitated out of water, and also the rocks' physical appearance, which includes niches where crystals grew.


Where did all the water go?

If there was once a lot of water on the surface of Mars, where did it all go? One possibility is that some of it evaporated, became water vapor in the atmosphere, and was then lost into space because the gravity pull of the planet couldn't hold on to it. Another possibility is that much of still exists on Mars, as ice. Both the poles of Mars contain water ice. And in 2002, scientists analyzing data from the Mars Observer spacecraft announced that they had found vast quantities of water ice in great swathes of the planet just below the surface.



1. Malin, Michael C., Kenneth S. Edgett, Liliya V. Posiolova, Shawn M. McColley, and Eldar Z. Noe Dobrea. "Present-Day Impact Cratering Rate and Contemporary Gully Activity on Mars." Science, Vol. 314. no. 5805, pp. 1573-1577, 2006.
2. Parker, T. J., R. S. Saunders, and D. M. Schneeberger. "Transitional Morphology in West Deuteronilus Mensae, Mars: Implications for Modification of the Lowland/Upland Boundary." Icarus, vol. 82, p. 111-145, 1989.
3. Clifford, S. M., and Parker, T. J. "The evolution of the Martian hydrosphere: Implications for the fate of a primordial ocean and the current state of the northern plains." Icarus, vol. 154: 40-79, 2001.