Image: Moon maps
On the left is an image of the moon showing brightness at shorter infrared wavelengths. On the right, the distribution of water-rich minerals (light blue) is shown around a small crater. Both water- and hydroxyl-rich materials were found to be associated with material ejected from the crater.
updated 10/6/2009 8:34:37 PM ET 2009-10-07T00:34:37

Many experts were shocked by the recent discovery of water on the moon, which was long thought to be bone-dry.

But not everyone was surprised.

Astrophysicist Arlin Crotts of Columbia University has been working for years on research that he says predicted this finding. In a paper he submitted recently to the Astrophysical Journal with his graduate student Cameron Hummels, Crotts hypothesizes the existence of widespread water on the lunar surface, and offers an idea for how it got there.

"I am predicting something that just happened, that nobody else was predicting," Crotts said. "I hope people recognize that this is a true prediction of the spatial distribution of water around the moon."

Until recently, many scientists thought the lunar surface was almost completely dry, and that shadowed craters near the poles offered the only chance for small stores of water. But new data from the NASA-built Moon Mineralogy Mapper (M3) on India's Chandrayaan-1 satellite, NASA's Cassini spacecraft and NASA's Deep Impact probe uncovered tantalizing evidence of water molecules all over the moon's surface. These findings were detailed in three papers in the Sept. 25 issue of the journal Science.

Some more details, especially about the possible water at the poles, are likely to come when NASA's LCROSS impactor slams into a crater on the moon's south pole Friday morning in search of signs of water.

Where did it come from?
The experts behind the new findings said they don't yet know the source of this water. According to one hypothesis, charged hydrogen ions carried from the sun to the moon by the solar wind could combine with oxygen on the moon to form water molecules. Another idea is that the water is left over from comets that have impacted the moon.

"There are many models out there," said Roger Clark of the U.S. Geological Survey in Denver, who is a team member for the Cassini spacecraft and a co-investigator for Chandrayaan-1. "Probably to some degree they all are in play. It's too early to tell."

But Crotts has a different idea in mind.

Previous research has uncovered some water trapped in minerals deep inside the moon, Crotts said. According to his model, this water is likely to travel up through fissures to the lunar surface along with other gases that are escaping the pressure of the moon's dense interior.

"We now know that there's water in the interior," Crotts told "There's no particular reason to think that it doesn't get out."

Buried water
One piece of evidence for interior water — a 2008 Nature study by Brown University's Alberto Saal and colleagues — identified water (between 260 and 745 parts per million, or ppm) in pebbles of hardened moon lava brought back by Apollo astronauts. Other work on similar samples by Francis McCubbin of the Carnegie Institution for Science in Washington D.C. also indicates the moon could harbor water beneath its surface.

Top 10 images of the world at nightWhile Crotts thinks those amounts are enough to produce the observed surface water, other experts are skeptical.

"I feel that it is highly unlikely that there are significant amounts of water remaining in the moon's interior at this time," said Darby Dyar of Mount Holyoke College in Massachusetts, who was a co-author on the recent Science papers announcing the surface water discovery. "The amounts of water found are at the parts per million level, and as such constitute only a very small amount of water as a resource."

Other scientists echo this thinking.

"The moon interior is believed to be very dry, with less water than what we observed on the surface," Olivier Groussin, a scientist at the Laboratoire d'Astrophysique de Marseille in France and another co-author on the Science papers, wrote in an e-mail. "Apollo samples indicate less than 50 ppm of water in the interior, while we detected about 1000 ppm on the surface."

Wet moon
However, Denton Ebel, curator of meteorites at the American Museum of Natural History in New York, said the trace amounts of interior moon water so far identified could be enough to produce the signature found at the surface.

"I think the amounts of water that are inferred for the lunar interior from the work of Alberto Saal and the work of Francis McCubbin, coupled with what we know about the lunar core, implies that degassing is a viable cause of the hydrogen signal that's observed," Ebel said in a phone interview.

"I think that [Crotts'] scenario of seepage — slow degassing — is consistent with the findings," Ebel said. "And I think it's more encouraging than the idea of hydrogen implantation by the solar wind. The bottom line is, he could turn out to be right."

Crott's paper outlining his hypothesis has been submitted to an academic journal, and is in the process of being peer-reviewed before possible publication. Some scientists are waiting to reserve judgment until then.

"I am delighted that scientists have been thinking along these lines, but we must wait to see if it holds up to the test of peer review," said Jim Green, director of the Planetary Science Division of the Science Mission Directorate at NASA Headquarters in Washington, D.C., another co-author on the Science papers.

To get to the bottom of the issue, more data will be needed, scientists say.

In fact, the signature of water seen on the surface could easily result from a combination of multiple processes, Crotts said, adding that his explanation might only account for some of the water on the surface.

To find out for sure, more lunar expeditions will be required, Crotts said.

'"We've got to have another polar orbiter mission, and it's got to have some instruments on it that study this question," he said.

© 2013 All rights reserved. More from


Discussion comments


Most active discussions

  1. votes comments
  2. votes comments
  3. votes comments
  4. votes comments