A new study from University of California Los Angeles shows that crystals found in moon rocks might have skewed water estimations on the lunar surface.

The reason for the overestimation is a mineral called apatite, which has been used as a marker to check water levels on the moon.

A new computer model developed by Jeremy Boyce of the UCLA Department of Earth, Planetary and Space Sciences and colleagues showed how apatite formed on early moon. The research found that the unusually hydrogen-rich apatite crystals in lunar rocks might not have been formed in a water-rich environment.

For several years, it was believed that the moon didn't have any water. But, studies on hydrogen-rich apatite obtained from lunar rocks suggested that the satellite might have had water in the past.

Apatite is a calcium-phosphate mineral. According to Boyce, the crystals incorporate hydrogen from water only when fluorine and chlorine have been mostly exhausted.

"Early-forming apatite is so fluorine-rich that it vacuums all the fluorine out of the magma, followed by chlorine," Boyce said in a news release. "Apatite that forms later doesn't see any fluorine or chlorine and becomes hydrogen-rich because it has no choice."

In other words, hydrogen-rich apatite in lunar rocks isn't a good indicator of water. The crystals just show that there wasn't much of fluorine and chlorine during crystallization.

The research is important not only because it shows that science might have to rethink apatite's role as an indicator for water, but also because the research could help resolve the mystery surrounding moon's origins.

A leading theory on moon formation is the "giant impact model," which says that an impact broke a piece of the earth some 4 billion years ago. If this model is correct, then the moon would have heated up and lighter elements such as hydrogen would have escaped into the space. Hydrogen is a key component of water, meaning that the lunar surface must have been dry.

Here's where the paradox lies. Lunar rocks do carry hydrogen, meaning that somehow water from earth might have remained on the moon despite the satellites fiery birth and development.

However, the amount of water in lunar surface might be far less than what apatite would have us believe.

"We had 40 years of believing in a dry moon, and now we have some evidence that the old dry model of the moon wasn't perfect," Boyce said. "However, we need to be cautious and look carefully at each piece of evidence before we decide that rocks on the moon are as wet as those on Earth."

The study is published in the journal Science and was supported by NASA Cosmochemistry grant and a NASA Lunar Advanced Science for Exploration Research grant.