Long journeys require human explorers to carry plenty of water. Astronauts need an average of 1.6 kilograms (0.4 gallon) of water to sustain each of them every day, as well as an additional 27 kilograms (7.2 gallons) for other purposes, according to NASA, and boosting even one of these kilograms into orbit costs $25,000. Given plans to send manned missions to the moon and, eventually, establish a base, it would clearly be far simpler if future explorers could fulfill their water needs there. Fortunately, early radar data hinted that water ice might exist in the permanent shadows of craters at the lunar poles, and the neutron spectrometer on board the Lunar Prospector in 1998 detected the telltale signature of hydrogen within the moon's surface. If that hydrogen is locked up in the form of H2O, then as much as 26 billion gallons of water could be frozen there.

To help settle the question, astronomer Donald Campbell of Cornell University and his team trained the world's most powerful radio observatory--the Arecibo Telescope in Puerto Rico--on the moon's south pole, particularly its large Shackleton Crater, searching for signs of water like those detected on Mercury and elsewhere. "Ice essentially lights up under radar. It preferentially reflects the light back at the radar like a highway sign," Campbell explains. "When we looked at the poles of the moon we did not see the same sort of signature."

But the astronomers did detect a polarization signature that could be indicative of water ice. Yet, the signature seemed to be coming from both the shadowed and lit portions of the craters. Because water ice, if it is there, could only exist in dark, cold recesses, the signature probably was caused by very rough or blocky terrain, the team argues in the paper presenting the result in the October 19 issue of Nature. "For reasons that we're not totally sure about, the same properties can arise from the scattering of rocky ejecta on the blocky terrain of young impact craters on the moon," Campbell notes. Such properties can be seen here on Earth in the volcanic flow from the San Francisco Peaks just outside of Flagstaff, Ariz.

The radar data obtained is the best to date, with resolution down to 20 meters and probing up to one meter into the moon's soil, even better than that obtained in the past by satellites with a more direct view from lunar orbit. But it is not the last word. "The definitive measurements have yet to be made," argues lunar scientist Paul Spudis of Johns Hopkins University. "It could be roughness or it could be ice, we just don't know."

NASA missions planned for launch a year from now--the Lunar Reconnaissance Orbiter and Lunar Crater Observation and Sensing Satellite (LCROSS)--should provide a clearer picture, and India's Chandrayaan-1 mission will carry a less sensitive radar system. "They will get a better view of the polar terrain than we can from the earth, and it will be interesting to see what they show up," Campbell says. "LCROSS is planned to impact into the moon and see if they can vaporize any water ice and then detect the vapor plume."

Regardless, the simple fact that hydrogen--and oxygen--in some form are on the moon makes a lunar base a simpler proposition. "If it turns out that [hydrogen] is locked in water and that is easy to get to and process, then all the better," notes NASA's lunar robotic precursor program manager Anthony Lavoie. "Even if it is not in water form, NASA will still be interested in determining if that form can be used."