Water in the atmosphere and on the surface?: Physical conditions on the surface of Mars and in its atmosphere create a regime where liquid water is generally unstable. Two important physical parameters influence this: (1) the small atmospheric pressure (2) the low temperature.

The Martian atmosphere is dry compared to the Earth's. Measurements from the Viking orbiter spacecraft showed that it contained the equivalent of 1-2 km3 of water ice. This compares to about 13,000 km3 of equivalent water in the Earth's atmosphere. Nevertheless, even with such small amounts of water vapor, sometimes the Martian atmosphere can "saturate" because it is so cold. When this happens ice clouds can form, fogs can freeze in low-lying areas, or thin films of frost can form on the surface.

The average atmospheric pressure on Mars is only about 6 millibars compared to the Earth's average pressure of 1013 millibars. The average surface temperature on Mars is about -60 deg C compared to the Earth's 15 deg C. Because the average atmospheric pressure is low and the temperature at the surface is generally below the freezing point of water, water generally only exists as ice or vapor. Thermodynamics (the science governing the relationship between heat and other forms of energy) does not allow liquid water. However, at certain locations and times on Mars, when the air pressure is high enough and the temperature is above freezing (greater than 0 deg C), liquid water is theoretically possible; but the rate of evaporation would probably be so great that liquid water, if present, would rapidly vaporize. Moreover, computer simulations show that the average surface temperature at a particular location gradually increases with the approach of summer; so by the time the melting point is reached, any surface ice would have turned to vapor. Liquid water on the surface is therefore possible only as a temporary phenomenon in places where there is some unexpected form of heating, e.g. sudden, near-surface geothermal activity.

Thus far we have only considered pure water. Dissolved salts are likely because the Martian soil is estimated to contain 8-25%wt salts. The presence of dissolved salts lowers the melting point and reduces the equilibrium vapor pressure considerably. Compared to pure water, this expands the regions on Mars where ice melting could occur and increases the total time such conditions might exist for liquid water. The northern low latitude region experience surface pressures higher than the triple point and temperatures above freezing. However, this region is likely to be dessicated because ice at these latitudes would migrate to the poles in a geologically short timescale. Basins in the southern hemisphere, such as Hellas and Argyre, are not low latitude sites and in theory may be good candidates for near-surface brines on present-day Mars.

A small number of high resolution pictures from NASA's Mars Global Surveyor Mars Orbiter Camera (MOC) appear to show gullies on cliffs and crater walls. This suggests that liquid water may have seeped onto the surface in the geologically recent past. The gully landforms are usually found on slopes facing away from noon sunlight, and mostly occur between latitudes 30° and 70° in both hemispheres. The relationship to sunlight and latitude may indicate that ice protects liquid water from evaporation until enough pressure builds for it to be released catastrophically. The comparative age of these features might indicate that some are still active today. Thus liquid water may presently exist in some areas at depths of less than 500 meters beneath the surface.

The image above shows the south-facing wall of an impact crater near 54.8°S, 342.5°W. This image (1.3 km wide by 2 km long) is illuminated from the upper left; north is toward the upper right. The right image is from the flank of the Mount St. Helens volcano in Washington state, USA (MOC Release MOC2-234, 22 June 2000; NASA/Malin Space Science Systems)

Water below the surface?: Theoretical calculations show that ground ice on Mars (ice buried in the soil) can exist very close to the surface only at high latitudes. Near the equator, ground ice can only survive if it is somehow isolated by an impervious rock layer to prevent it gradually turning to vapor and being lost to the atmosphere. As you go down through the ground at a particular location beyond tens of meters depth, the temperature will increase, as it does on the Earth, because the interior of the planet is hot. At a certain depth, called the melting isotherm depth, the temperature will exceed 0 deg C, the freezing point of water. Below this depth, liquid ground water could theoretically exist. If we take 0 deg C as our freezing point, the melting isotherm is calculated to be 1 to 8 km below the surface. Actually, the freezing point could be somewhat lower if salts are dissolved in the water, leading to a smaller depth for the melting isotherm. A possible salt that may be dissolved is sodium chloride - i.e. common table salt. It may also be possible that geothermal activity at certain locations could lead to liquid water closer to the surface - but no "hot spots" have yet been identified by spacecraft missions (See "Are the Martian volcanoes extinct?").

The latitudinal and depth distribution for ground ice on Mars (from a climate calculation done for when Mars was at low obliquity (approx 15 degrees) but essentially similar to how it should be today (when obliquity is at 25 degrees). [Obliquity is the axial tilt of a planet, i.e. the angle between its axis of rotation and the pole of its orbit, responsible for the seasons --- for Mars, obliquity oscillates over timescales of 100,000 and a million years]. Black areas are at temperatures below freezing point where ice could exist if enough water is present to completely fill the soil pore space - otherwise the bottom of the ice is shallower than depicted. The gray area is where lesser amounts of ice (like pore frost) could persist if it is replenished by upward diffusion from a source below (e.g. ground water).