The neutron and gamma-ray spectrometers on NASA’s Mars Odyssey discovered that water ice lies at shallow depths from the polar regions down to latitudes of about 55° north and south. Images of very recent craters by the HiRISE camera on the Mars Reconnaissance Orbiter have revealed fresh ice exposed in crater bottoms at latitudes as low as 43° north.
ICE IN RETREAT. A small crater, just 6 meters (20 feet) wide, shows ice in the bottom when the HiRISE camera imaged it in October 2008 (left), sometime soon after an impact made the crater. By January 2009 (right), much of the ice had disappeared. Left image is from HiRISE image PSP_010440_2235, right image is from ESP_011574_2235. Images: NASA/JPL-Caltech/University of Arizona
A paper in Icarus by Norbert Schorghofer (University of Hawaii) and Francois Forget (Université Paris) presents results of modeling how ground ice comes and goes on Mars. They argue that the ice-exposing impacts point to an underground ice sheet that’s still retreating from the last glacial maximum.
Models predict that the size of the subsurface ice layers in both hemispheres has changed over the past few million years. These shifts are driven by changes in the Martian orbit and shifts in the orientation of the planet’s rotation axis.
“Subsurface ice can be emplaced in two ways,” the researchers explain. “In one method, snowfall during a past climate period when the axis tilt was different may have led to the formation of a perennial snow cover that subsequently densified.” When this ice retreated, any dust caught in it would remain as a growing lag deposit of debris, eventually burying the ice.
The second mechanism, they note, is uncommon on Earth: underground ice is deposited directly from water vapor. “This ice fills the available voids between soil grains and is thus called pore ice or interstitial ice.”
Their modeling of subsurface ice and its interaction with the atmosphere leads them to three conclusions. First, when ice in soil forms from atmospheric vapor, it can grow gradually over a range of depths below an ice table. It can also grow if the ice table moves to shallower depths.
Second, at the Phoenix landing site (68° north), calculations predict three layers: dry soil, ice-cemented soil, and an underlying ice sheet. But the layer of ice-cemented soil may be only millimeters thick, thinner than expected.
Finally, the ice-exposing crater impacts at 43° north lie slightly closer to the equator than expected for subsurface ice in equilibrium. A likely explanation is a recent massive underground ice sheet that is still retreating because it has not yet reached equilibrium with the present day atmosphere and climate.