Inside NASA's audacious plan to build a nuclear-powered moon base

Over the past 10 days, four astronauts flew around the moon, travelled further into space than any humans in history, and will soon splash down into the welcoming waters of the Pacific Ocean. The spectacular success of NASA’s Artemis II mission has now set the stage for the U.S.’s return to the lunar surface, currently planned for early 2028—which, in turn, readies humanity to irreversibly transform the moon into a permanently inhabited location.

Realizing this vision of the moon, as a bustling off-world outpost, may not be as far off into the future as you may think. Just prior to Artemis II’s launch, NASA Administrator Jared Isaacman announced a revamped vision for the Artemis program, including a more aggressive launch schedule, and plans to kickstart the construction of a base on the lunar south pole in late 2028. And while the Artemis program begins sending astronauts to moon, a new space race is heating up. Private spaceflight companies, foreign governments, and international agencies will fly their own uncrewed missions to the surface—both to support NASA’s efforts and to start prospecting work of their own.

(Follow along with National Geographic’s continued coverage of Artemis II here.)

Within just a few years, there may be one U.S.-led mission sent to the moon every single month. “They’re talking about such a high cadence,” says Philip Metzger, an expert on spaceflight engineering at the Florida Space Institute. “There could be a lot of expeditionary missions to all over the moon.”

But first, many things must go right, and the short timelines proposed by NASA’s leadership are certainly ambitious, considering the agency’s recent history of missing Artemis-related deadlines; Artemis II was originally supposed to launch in 2024.

Still, over the past month, many in the broader space science community feel that the race to return to the moon has gained real momentum. “It was like an opportunity grenade went off,” says Tim Crain, a co-founder of Intuitive Machines, a Houston-based space exploration company currently focusing on developing lunar technology, including lunar surface transportation and cargo landers.

If all of NASA’s audacious plans do come to pass, the moon might become a very unfamiliar place—a hub of industry and science, crisscrossed by a fleet of moon rovers, fueled by nuclear power, and the launching point for even farther cosmic adventures to come.

Next stop: the Lunar south pole 

The Artemis program is ultimately about one thing: “establishing that enduring presence” on the moon, says Lori Glaze, NASA’s Moon to Mars program manager.

Artemis III, currently slated for 2027, will test spaceflight operations in low-Earth orbit: a crew will pilot the Orion spacecraft—the same type used by the Artemis II astronauts—and attempt to dock with landers sourced from SpaceX or Blue Origin (or both at the same time). These landers are designed to shepherd the crew of future Artemis missions to the lunar surface, including Artemis IV (scheduled for early 2028) and Artemis V (late 2028).

All those missions will enable NASA to start on its ultimate goal: building a base, which will probably be situated somewhere in the lunar south pole region—a benighted swath of land, with some incredibly inhospitable conditions. It is essentially a volcanic desert, one with temperature swings of hundreds of degrees, abrasive lunar dust, and a near-constant shower of cosmic radiation from the jet-black sky above. But what makes this deathtrap so desirable is that it’s home to reserves of frozen water.

The most promising locations, based largely on orbital data, are in craters that haven’t seen sunlight in several billion years. “That’s where we believe it’s cold enough that you can keep the ice in a frozen state for very long periods of time,” says Glaze. It’s open to debate as to how much ice is actually present there, and in what form—but some of the lower estimates are still sufficient to get experts excited.

George Sowers, a mechanical engineer at the Colorado School of Mines, previously told National Geographic that “water is the oil of space,” important for establishing an energy resource for a lunar economy. If you expose water to electrical currents, you split it, generating oxygen and hydrogen; the former can be used to create breathable air, while both can be combined to make rocket fuel. That water can also be used to hydrate both astronauts and crops. Altogether, this would make the moon base—and any future crewed parts of the moon—self-sufficient, and not reliant on staggeringly expensive deliveries from Earth. 

The moon, then, won’t just become a collection of new flags in the dust. A base will turn it into a home for astronauts and a steppingstone to crewed missions to the red planet next door. Launching rockets from the moon is, thanks to a lack of atmosphere and low gravity, far cheaper than doing so from Earth. Or as Clive Neal, a lunar geoscientist at the University of Notre Dame, puts it, "this is really a blueprint of how you need to do things in order to make Mars a reality.”

Building a castle on the moon

Establishing the very first lunar bastion will occur in a few distinct phases.

Step one, starting with the Artemis V mission, will be gradual experimentation: using mostly uncrewed missions to test out basic technological elements—from power generation to communications relays—to make sure base building can be done effectively and safely. During this phase, the number of launches to the moon will begin to rise. Then, with the help of regular robotic and astronaut visitations, the foundations of the base will be set down during the second phase. At this stage, NASA describes it as “semi-habitable infrastructure,” although it’s not yet clear what this means, precisely. (Early ideas include inflatable shelters, or covering a habitat with lunar soil to help insulate astronauts from radiation.)

Finally, there’s the third phase, when frequent, heavy cargo deliveries and significant contributions from NASA’s partner space agencies will turn a small, periodically homed fortress into a permanent base, one that is always stationed by a crew, like the International Space Station is today. The idea is that astronauts will spend “a few days to a couple of weeks on the surface, and then build up to something longer—maybe a month, maybe a couple of months,” says Glaze.

The original base will primarily serve as a scientific research outpost, one that can be added to in a modular manner—plugging in new sections as needed, like lunar Lego. Here on Earth, researchers are already hard at work perfecting a method to transform lunar regolith into more earthlike soil, capable of growing specially designed crops. If they succeed, a lunar greenhouse could become a reality.  Another key piece of early architecture will be a landing pad, which must be built far enough from the main base to stop corrosive rocks and dust from smashing into it each time a spacecraft touches down.

Driving around the moon

But don’t picture all lunar development as looking like a city, with a dense central hub and expanding outwards from there. “We’re going to establish hubs remotely,” says Crain: small, far-flung scientific outposts studying water-rich deposits within remote craters, for example, or power plants installed a safe distance from the crew.

This sprawling infrastructure won’t be possible without the use of Lunar Terrain Vehicles, or LTVs: rugged buggies equipped with manipulating arms and scientific tools that can be piloted by astronauts, driven remotely from Earth, or left to autonomously drive themselves across the moon.

Three companies (including Intuitive Machines) are currently bidding to win a multi-billion-dollar contract to be the official Artemis-era LTV. But all three aspire to send their LTVs to the moon regardless of the competition’s outcome, whether it be with NASA or private spaceflight endeavors. “There’s going to be a lot of rovers needed just to carry the logistics,” says NASA’s Glaze.

And often, astronauts won’t be using them directly. Instead, while astronauts carry out critical tasks, the LTVs will roam about and perform a lot of the work on their own. “The more autonomy that we have, the better we can spend the limited time we have with the crew while they’re on the surface,” says Glaze. And if they aren’t building outposts, these LTVs will roam into the distance, using their water-divining tech to scout out additional lunar oases.

The near-future moon will also be watched over by an array of yet-to-be-launched satellites. Earth’s companion will require its own versions of our planet’s GPS and telecommunications networks in the sky. The first version will either be launched by a private company, or by a space agency; Europe’s Moonlight project, for example, hopes to deploy a small constellation of lunar satellites in the next few years. Lengthy base-building, particularly using LTVs, will require a network like this to operate.

It won’t be long before the first-ever lunar factory is also established, says Crain. Using a mixture of resources sent from Earth and those from the terrain around it, he imagines it will manufacture the most rudimentary part of any off-world base: panels—flat surfaces you can turn into walls, radiator panels, solar panels, or landing pads. It’s not difficult to picture robots being left alone to make bricks out of the lunar soil for LTVs to scoop up and stack together.

Space Age prospectors 

If you’re on the lunar nearside—the side that always faces Earth—solar power can provide a decent amount of energy needed to fuel simple scientific investigations. But a moon base and its disparate collection of hubs and beacons—augmented by dozens of LTVs zipping about—won’t work on solar power alone. This is especially true at the lunar south pole, where nights can last for 14 agonizingly frigid Earth-days.

Experts agree that the only solution to this is to use nuclear fission power. “Nuclear clearly solves the long lunar night problem, and so that’s the big draw,” says Metzger. A tiny mass of nuclear fuel could power a small lunar base for a couple of decades.

To start, last year, NASA said that it was aiming to deploy a 100-kilowatt nuclear reactor—something 10,000 times less powerful than a typical nuclear power plant, which could fit into a small car—to the lunar surface.

Designing a nuclear reactor that could work on the moon is a novel engineering challenge. The U.S. has only deployed one nuclear reactor to space before—an Earth-orbiting satellite named SNAP-10A—way back in 1965, which ran for 43 days before it stopped working. But just prior to Artemis II’s launch, Isaacman also announced that the agency is building an experimental nuclear reactor-powered spacecraft, one that will be launched from the Earth towards Mars by the end of 2028. Whether the experimental spacecraft works or malfunctions, it will provide a wealth of information to aid development of a lunar power plant. (Launching a nuclear reactor into space, although not without its risks, is also not as dangerous as it sounds: the nuclear fuel itself isn’t dangerously radioactive, but the waste produced during nuclear fission is—which is why the reactor won’t be turned on until it’s left Earth.)

For the foreseeable future, “small-scale microreactors ready for operation on the moon are not so far-fetched,” says Katy Huff, an associate professor in the Department of Nuclear, Plasma, and Radiological Engineering at the University of Illinois at Urbana-Champaign. But eventually, somewhat larger nuclear power plants would be in demand—and not just to support a cozy domicile for astronauts and an ever-expanding suite of scientific investigations. “Eventually, you get to a power level in the hundreds of kilowatts,” says Rian Bahran, the Deputy Assistant Secretary for Nuclear Reactors at the U.S. Department of Energy, “where you can do industrial processes.”

The creation of a new lunar economy

With all that power established on the moon, the lunar surface will be open for business. Glaze refers to building “a commercial lunar ecosystem.” Although at this point the exact shape of such an economy is very hazy, the commercial mining of two lunar resources is a common theme among space experts.

The first, and most important, is water—which, as we know, can allow lunar operations to proceed independent of supply runs from Earth. The second is helium-3: a rarity on Earth, but something the Moon has in large quantities. As the moon lacks a magnetic field, it has been showered by the sun’s radiation for eons. While this means that astronauts will require their shelters to be hardened against this cosmic precipitation, it also means that the surface has been seasoned with that strange form of helium, which is carried to the surface by the solar wind (i.e., the stream of charged particles that burst out from the sun’s atmosphere).

Evidence suggests that, unlike water-ice, helium-3 is concentrated in areas that have enjoyed prolonged sunlight exposure—particularly closer to the equator, including on the moon’s always-visible nearside.

At present, humanity doesn’t have much use for the substance. But helium-3 has excellent cooling properties, which means it could be used in the energy-intensive data centers and quantum supercomputers of the future. Helium-3 would also be an effective fuel source for nuclear fusion. Compared to preexisting fission, fusion is a far more efficient, persistent and cleaner version of nuclear energy. It has yet to overcome some significant engineering hurdles, but if scientists work out how to harness the power of a star to electrify the world, helium-3 will be fiercely sought after.

Helium-3 could be incredibly difficult to mine out of the lunar soil. But investors are bullish on its worth, and just one literal handful of the stuff is priced at around $20 million. A few decades from now, on a cloudless night on Earth, people might see the faint, speckled glow of the lights of several nuclear-powered helium-3 factories on the lunar nearside.

A new space race 

If astronauts find that there really isn’t that much accessible water on the moon, the lunar economy might splutter and die. And even with the best available resources, challenges will remain—like protecting a moon base from meteors. Our planet’s atmosphere filters out smaller asteroids, but the moon has no natural shield. That means impactors the size of houses—which are pretty commonplace—can carve out craters in the lunar surface big enough to obliterate bases. Planetary defense technologies, which currently only apply to Earth, will need to be deployed to the moon.

But if a lunar base survives and thrives, “the space economy could grow to be as large as the terrestrial economy within 50 years,” says Metzger.

Which means if NASA’s plans get realized, the moon could be transformed. But the U.S. is not alone in its ambitions. China may attempt its first crewed landing on the nearside of the Moon, but it also aims to eventually set up shop on the water-rich south pole; one or both could happen by the end of the decade. It also has an agreement with Russia to set up their own nuclear reactor on the moon, one that will power a jointly-operated Sino-Russian lunar base. China has been sending its robotic spacecraft back and forth to the moon over the past few years; this has in part been to obtain fresh samples of lunar rock to study, but each mission is also being used to practice flight maneuvers and technology that could be deployed in a crewed mission.

The U.S.-led Artemis program, then, has a direct rival that’s quickly gathering steam. And winning this space race will yield long-term prizes. 

Arguably, much depends on those caches of lunar water, whose exact quantities are unknown. “If the sites are very plentiful, then it could defuse the geopolitical tensions,” says Metzger. “If those resources are limited, there might be an aspect of urgency to secure some of the better locations before adversarial countries do.”

In other words: you want to be the first to do anything on the moon, to set the ground rules going forward. That applies to everything, from the use of nuclear power to mining operations. “The policies for the next 100 years are going to be decided in the next 10 to 20 years, so you want to be there operating on the moon to establish those policies,” says Metzger. 

Some experts, particularly those in Europe, worry that in the rush to claim spots on the lunar south pole, the U.S.’s international partners will be left behind. In the most optimistic scenario, “my dream would be that we use the moon as a science base, like we have in Antarctica,” Patrick Michel, a planetary scientist at the University of Côte d’Azur in France, says. The moon could be a place where multiple nations peacefully decide to share their resources, “at the service of exploration and science.” 

If Artemis is one small step toward that goal, then the next giant leap may be to Mars itself. Can humanity become interplanetary someday? “The moon is the best proving ground,” says Glaze.

On April 6, after 40 minutes of radio silence, the Artemis II crew reemerged from behind the moon. Before them, Earth and its natural satellite lingered, together, in the void. Looking on from the Orion capsule, astronaut and mission specialist Christina Koch was feeling sanguine about their futures.

“We will explore; we will build. We will build ships. We will visit again. We will construct science outposts. We will drive rovers. We will do radio astronomy. We will found companies. We will bolster industry. We will inspire but ultimately, we will always choose Earth,” she said. “We will always choose each other.”