Ice unveiled. A meteorite impact has revealed nearly pure ice (white dot in yellow-brown smudge)--possibly left from an earlier ice age--beneath a thin layer of martian soil.

NASA/JPL/University of Arizona

Blasting for Ice on Mars

The Phoenix lander has given scientists a close look at the ice in one spot high in the martian arctic, but researchers have also been surveying fresh craters across the planet for signs of frozen water. Now two teams have reported ice that is relatively pure, far purer than expected on a dusty, dirty planet. The finds lend support to a scenario of an ancient "iceball Mars" in which ice encased much of the planet.

Sites of small meteorite impacts just a few months old are most likely to still show traces of subsurface ice. Two cameras on board the Mars Reconnaissance Orbiter (MRO), which first began orbiting the Red Planet in 2006, are working in tandem to ferret out these craters. As planetary scientist Shane Byrne of the University of Arizona (UA), Tucson, and his colleagues reported last month at the Lunar and Planetary Science Conference in The Woodlands, Texas, MRO's broad-scale Context Camera has been finding the dark smudges of impact debris that weren't there as recently as a month earlier.

MRO's High Resolution Imaging Science Experiment (HiRISE) has imaged in detail the new craters. Spanning northern latitudes of 43° to 56° (Phoenix landed at 68°N), five new craters were a few meters in diameter, a few decimeters deep, and floored with a bright, white material that looked like clean ice. But theoreticians were not expecting anything like pure ice. They had assumed that atmospheric water vapor had seeped into high-latitude martian soil and frozen between soil particles, forming a half-ice, half-soil mixture. Theory aside, MRO's spectrometer clearly identified the bright ejecta around one larger crater as water ice.

In a second talk, HiRISE team members Colin Dundas, Byrne, and Alfred McEwen of UA showed how they inferred that the newly exposed ice really was nearly pure water. HiRISE repeatedly imaged a crater that was no more than a month old as the ice sublimated away, leaving behind any dust and dirt it contained and obscuring the remaining ice. But the ice patch did not fade away until 3 months after formation; dirty ice at the Phoenix landing site formed an obscuring dirt layer within just 2 days. That suggests that the impact-revealed ice is "relatively pure," said Dundas.

"It's a very promising discovery," says Norbert Schörghofer of the University of Hawaii, Manoa, who models ice in martian soils. "What it means, we have to wait a little longer." But the clean ice "didn't come as a huge surprise to us," says planetary geologist James Head III of Brown University. Drawing on odd, icy-looking landforms in earlier images, he and colleagues proposed in 2003 that in the geologic past, snow-fed glaciers and ice fields had covered much of the lower latitudes of Mars at the expense of the polar ice caps (Science, 11 April 2003, p. 234). "That ice is still there," he says, beneath a thin layer of soil.