The answer is either that it has escaped from the planet or is hidden somewhere on or near its surface. That water has escaped would be in agreement with the low concentration of water in the Martian atmosphere.

There are, however, ways to hide water in ice form for hundreds of millions of years as a substrate beneath a sufficiently thick insulating surface layer. Such a surface layer, sediment or lava flow, would prevent its transport to the atmosphere.

Ice on Arsia Mons.

The volcano-tectonic setting of Arsia Mons and its neighbours, Pavonis and Ascreus Mons, is analogous to numerous terrestrial volcanoes. From a central caldera a fissure swarm extends for hundreds of kilometres and carries magma from a chamber beneath the caldera to distal parts of the volcano outside the caldera rim.

Recently, it was proposed, on the basis of a morphological interpretation of Viking Orbiter images, that a thick ice sheet covers the Arsia Mons volcano in the Tharsis province on Mars, i.e. almost on the equator (Geology, March 1999, v. 27; no. 3; p. 231-234).

This conclusion is reached by examining deep canyons that cut into the fan areas of Arsia Mons and almost reach into the caldera area. These canyons are interconnected to circular vent-like structures that are interpreted as openings where ice has been melted above eruptive volcanic sites.

These vent-like openings lack a collar of ash normally associated with eruptive vents. The close association of a fissure swarm, large scale canyons, numerous vent-like structures and a broad fan area concur with a genetic relationship between these features.

Thus the conclusions are;

• meltwater has been the erosive agent that formed the canyons,
• the openings are formed as the result of melting above a subglacial eruptive site,
• debris flows extend to the distal fan areas, a distance of about 1000 km, appears reasonable.

Subglacial volcanism.

When volcanism takes place beneath ice the products accumulate vertically and the hot lava melts enclosing ice. The meltwater will eventually leak out of the system and this occurs through catastrophic flooding of a high standing water mass. Such catastrophic floods are well known and frequent in Iceland and referred to as j�kulhlaups. One such violent j�kulhlaup occurred in 1996 from beneath the Vatnajokull glacier with discharge volumes of over 50.000 m3/sec.

It is typical for circular depressions to form within glacier ice in response to melting from below. In Iceland the accumulation of seasonal new ice "heals" such depressions, whereas on Mars they would be left as permanent "holes" in the ice shield in absence a water rich atmosphere.

Above:This photo shows conditions when the hot magma beneath the ice is accumulating toward the surface and causing the ice above to subside and form circular depressions, just prior to explosive activity. Image by Ragnar Th. Sigurdsson Arctic Images

The response by some scientists to the idea of substantial ice volumes near the Martian equator has been one of scepticism. Bruce Jakosky of the University of Colorado at Boulder regards it problematic to accumulate such large ice volumes in one location (New Scientist, March 13th, 1999).

Since Jakosky made his comment Helgason has constrained his model and estimated the volume of ice that he believes is to be found in the Tharsis regime (5th International Conference on Mars, Pasadena, July 18-23, 1999).

He assumes that the vast aureole deposits by Olympus Mons are formed through the failure of an ice layer underneath. He argues that at one time ice covered an area extending from Olympus Mons in the northwest to beyond Arsia Mons in the southeast. By assuming a 500-m-thick ice layer equally distributed over Tharsis, as much as 5 million cubic kilometres may still be present there.

The volcano-tectonic setting of Arsia Mons and its neighbours, Pavonis and Ascreus Mons, that are some 26 km high, is analogous to numerous terrestrial volcanoes. From a central caldera a fissure swarm extends for long distances, sometimes hundreds of kilometres.

The fissure swarm carries magma from the caldera to distal parts of the volcano beyond the caldera walls. Upon reaching an ice-rock contact outside the caldera the ice melting process begins.

The meltwater can carry a high concentration of rapidly quenched magma, that will eventually burst out on the surface to rest in a distal location away from the higher reaches of the volcano. The scarcity of vent-like openings within the caldera suggests that this is where the ice thickness reaches a maximum.

Microwave probing of Mars.

Strong support for ice on Arsia Mons has come from microwave probing of Mars by a team lead by professor Duane Muhleman of the California Institute of Technology. His team found nine regions that were anomalous or more than three times stronger than the mean disk, the strongest being the Residual South Polar Ice Cap with a reflectivity of 100%.

Nearly all of the other 8 regions were in the Tharsis volcanic regime. Three of them were on or near the caldera of the stratovolcanoes Arsia, Pavonis and Ascraeus Mons.

A new $4 million NASA research investigation headed by professor Don Gurnett of the University of Iowa was recently contracted to search for Martian water through the use of low-frequency radar signal that will penetrate to a depth of five kilometres.

This study, which is part of the European Space Agency's (ESA) Mars Express spacecraft to be launched in 2003, may be pivotal in the search for the lost Martian water. If technically successful, the project is likely to provide a wealth of information on the ideas presented above for the Tharsis water/ice budget.

This article was contributed by
Johann Helgason
Ekra Geological Consulting
Thorsgata 24, 101 Reykjavik Iceland