Luke Sharrett / Getty Images

Ten years ago, a rocket slammed into the moon.

The impact sent a plume of lunar material from the moon’s south pole flying out into space. For a few minutes, the spacecraft that had unleashed the rocket coasted through the mist, its instruments absorbing as much data as they could. Amid the molecules of methane, ammonia, carbon dioxide, and other compounds, the spacecraft detected something wonderfully familiar: water.

Not liquid water, but grains of water ice. The discovery helped reshape our understanding of Earth’s satellite. Though scientists had long believed that the moon was quite dry, they had begun to harbor suspicions that water might lurk somewhere in its shadowy regions. The excavated material showed them they were right to wonder. It wasn’t much, but it was enough to suggest there was a lot more.

This is where NASA wants to go next, to craters along the moon’s south pole untouched by sunlight. Jim Bridenstine, the agency’s administrator, brings up water ice almost every time he talks about the Artemis program, the Trump administration’s effort to return astronauts to the moon in the next five years. The hope is that future spacefarers could mine the ice as a resource for their moon bases. “We know that there’s hundreds of millions of tons of water ice on the surface of the moon,” Bridenstine says. Sometimes he says there’s hundreds of billions of tons.

But Bridenstine doesn’t know that—not for sure. No one does.


Out in the cosmos, water is actually everywhere, usually in the form of ice. Its signature has been found beneath the surface of Mars, in the atmospheres of exoplanets, and inside dusty interstellar clouds. It is not strange to find water beyond Earth, though our planet, cloaked in oceans, certainly wears it best.

The moon had a dry reputation from the very beginning. According to the leading theory, it formed from the debris of a collision between Earth and a Mars-size object about 4.5 billion years ago. The impact was so fiery that scientists suspected that any water, whether it came from Earth or the mystery object, would have boiled away for good.

But in the late 1990s, an orbiting spacecraft flew over the poles and detected an abundance of hydrogen, which, when combined with oxygen, forms water. The hydrogen felt like a bread crumb beckoning scientists to follow. The moon’s axis has a very small tilt, which means sunlight never reaches some polar regions. The cold and dark conditions, some scientists predicted, could protect any water ice from the sun’s destructive glare.

Scientists tested this theory, a decade later, with force. The Lunar Crater Observation and Sensing Satellite (LCROSS) arrived at the moon in 2009 with a rocket booster, now empty, that had helped launch it into space. Hurled down to the surface, the projectile exhumed those grains of pure water ice, hidden in darkness for perhaps billions of years, and lofted them into the light.

Other experiments around that time provided more evidence of lunar water. Scientists detected hints of water in Apollo samples of volcanic glass, a remnant of the moon’s fiery beginnings. One astronomer, poring over images of the south pole, noticed that some spots on the surface shared similarities with minerals that require water to form. Even Cassini, a mission bound for Saturn, caught something. Cassini had turned toward the moon—a nice dry target—to calibrate its instruments on its way out and picked up some contamination that later turned out to be a signal for water. It was becoming very difficult to ignore the new story of the moon.

Today, scientists believe the moon harbors water inside and out. There is likely an ancient reservoir deep in its interior, and when the moon was young and molten, water escaped through volcanoes, froze in the vacuum of space, and rained down on the surface in beads of glass, the kind that astronauts later collected. The glass is found beyond the moon’s polar regions, but the trapped water is difficult to extract, says Anthony Colaprete, a planetary scientist at NASA’s Ames Research Center who led the LCROSS mission. “The glass needs to be heated to very high temperatures to drive the water out,” Colaprete says.

The surface, even the parts that receive sunlight, is sprinkled with traces of water. This water arises when charged hydrogen atoms from the sun strike the lunar regolith, split oxygen bonds in the soil, and join with them to produce water. The final product is only a few molecules deep, though—no use for thirsty astronauts.

The stuff that future astronauts really want lies in the shadows, inside craters where the sun never shines. Like Earth, the moon was probably bombarded with water-bearing asteroids and comets in its early days. Without an atmosphere to disintegrate them, the objects smashed into bits on the surface. Particles of water ice, newly exposed, scattered. The ice that drifted into darkened craters, or even small shadows cast by boulders, survived. “Once a molecule got into one of these areas, it could never get out,” says Carle Pieters, a planetary scientist at Brown University who oversaw a mineralogy instrument on an Indian robotic mission to the moon in 2008.

NASA and commercial space companies have set their sights on these mysterious, enduring reservoirs at the south pole, and they have based their estimates of the supply on the past decade of exploration. Bridenstine’s estimate, according to his office, likely comes from NASA’s chief scientist, Jim Green. Several scientists tell me the estimate is not outlandish, but they stress that it is merely a range. “If you take 10 scientists in a room, you get multiple answers,” says Thomas Zurbuchen, the NASA associate administrator who leads the agency’s science programs.

So what if astronauts were on the moon right now, and they could rappel into one of these craters and see the water ice for themselves? They would probably find something like “a dirty snowbank,” says Lindy Elkins-Tanton, a planetary scientist at Arizona State University. “It’s going to be a mess,” she says. “It’s going to be water and sulfur oxides and ammonia, and it’s going to be shards of rock and glass from the moon, and then it’s going to be a lot of organic materials.”

Astronauts would have to figure out how to extract lunar dust, metals, and other materials that could be harmful if consumed. “There’s been a little bit of an assumption that we can use the water-purification systems that we use here on Earth,” Elkins-Tanton says. “The water we’re going to find on the moon is not like any water we’ve ever had to process on Earth.”

To pin down the nature of water ice on the moon, scientists need fresh data from new missions. NASA wants to send landers and rovers to probe the silvery terrain for answers.

But even the most advanced rovers could miss signs of water, says Rick Elphic, a planetary scientist at Ames who is working on science instruments for the moon program. Spacecraft observations have shown that while total darkness provides the right conditions for water ice, it doesn’t guarantee the presence of water ice. And direct hits from meteors can scramble the shadowed regions that do have ice.

“As time goes by, impacts and micrometeoroid particles would continually erode, remove, bury, and redistribute the icy material,” Elphic says. “Occasional impacts would act like a hole punch, removing areas of ice-bearing material.”

The cosmic hole-punching, over the course of hundreds of millions of years, could mold a landscape of icy islands, with barren rock in between. “These ice-bearing islands might be few and far between, and very irregularly shaped,” Elphic says. A NASA rover could drive for miles and not detect a single crystal.

In a world where funding didn’t matter and physics cooperated, scientists would send machines crawling all over the lunar surface, from the equator to the poles. But moon missions are difficult and expensive, even without any people on board. Just last month, an Indian spacecraft designed to explore water ice near the south pole stopped transmitting right before touchdown, and hasn’t been heard from since.

For now, scientists must contend with what they have: hints and traces, the decade-old data from a plume of disentombed particles. Until rovers—or people—start roaming the bottom of the moon with drills and chisels, the basins of water ice that Bridenstine preaches about remain in the realm of daydreams.

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.