The unequivocal and unexpected discovery of widespread water in the uppermost layer of the moon's surface means a whole new set of challenges for engineers working on methods for future astronauts to extract lunar resources.

The amount could be as much as 1,000 water molecule parts-per-million, which could yield up to 32 ounces of water, or one quart, for each ton of the top layer of lunar regolith. Getting that water out poses a new challenge for lunar exploration engineers, who have been focused on techniques for retrieving water ice from permanently shadowed craters at the lunar poles.

"The process for harvesting it is completely different from what they were intending with the permanently shadowed areas," says Carle Pieters, principal investigator for NASA's Moon Mineralogy Mapper (M3) instrument, which sparked the water findings. "So that's a brand new challenge that is now on their table."

Most previous searches for lunar water have focused on the permanently shadowed craters at the moon's poles as likely locations where water ice could accumulate. But data from M3 - an imaging spectrometer carried aboard India's now-silent Chandryaan-1 lunar probe - has directly detected water and hydroxyl (OH) molecules in the upper few millimeters of the lunar regolith in sunlit areas, with the strongest readings at the moon's cooler, higher latitudes.

According to M3 co-investigator Roger Clark, "this was thought to be impossible - to have water on the surface of the moon in hot sunlight."

Pieters and her colleagues spent several months debating and rechecking the results, eventually finding confirmation in older calibration data sets gathered by spectrometers on NASA's Cassini probe during a flyby in 1999, and from the EPOXI (formerly Deep Impact) spacecraft in 2008. "There's no question that it's clearly not one of the instruments having a calibration problem," Pieters says.

How the water got there is still a subject of conjecture, with theories ranging from cometary impacts to the interaction of the solar wind with the lunar surface.

Scientists found that the amount of OH/H20 detected varied inversely with the amount of sunlight falling on the soil. When it was cooler and dimmer, the reading was higher, and when it was warmer and brighter, the reading was less.

This temporal variation points to the possibility of water and hydroxyl being periodically created and lost in a regular cycle.

If that water somehow becomes mobile, it also could provide a mechanism for ice to accumulate in lunar craters. NASA hopes to verify such deposits when its Lunar Crater Observations and Sensing Satellite (LCROSS) impacts crater Cabeus A near the moon's south pole on Oct. 9.

M3 managed to map 90 percent of the lunar surface at low resolution before Chandrayaan-1 stopped transmitting signals from lunar orbit on Aug. 29, having completed 10 months of a planned two-year mission.

Had the mission continued, M3 would now be gathering high-resolution data.

Artist's concept of Chandryaan-1: ISRO