A version of this story appeared in Science, Vol 373, Issue 6560.
Next year, NASA’s first mission to the lunar surface in 50 years will be run from an unlikely place: a low-slung building wedged between fast-food joints just off the Ohio River.
This unassuming former data center in Pittsburgh is the new home of Astrobotic, one of a few startups that NASA has selected to ferry scientific instruments to the lunar surface as part of the agency’s $2.6 billion Commercial Lunar Payload Services (CLPS) program. Starting next year, CLPS landers will reach the Moon’s surface at least twice a year, the agency hopes. It’s an astonishing pace after a decadeslong drought for U.S. science on the lunar surface, says Brett Denevi, a planetary scientist at Johns Hopkins University’s Applied Physics Laboratory. “We’ve been talking about this so long. It’s almost shocking to see it happen.”
Even without a single launch, CLPS has already radically changed the face of lunar science, says Stuart Bale, a space scientist at the University of California, Berkeley. “It feels like the Wild West. It’s lean, mean, and cheap.” Instruments that are less proven can be developed, flown, and tried out in a matter of a few years. They promise to bring a bounty of new science: maps of the subsurface water recently detected from lunar orbit, the first probes of the lunar interior since the Apollo landings, visits to mysterious magnetic anomalies, and unprecedented views of Earth’s magnetic field and deep space.
One day in June, Astrobotic was buzzing during a visit by its ultimate VIP: Thomas Zurbuchen, head of NASA science. Armed with the Swiss accent of his birth and an alarming directness, he had arrived at Astrobotic’s headquarters only to find much of it a construction site—work had been delayed by the pandemic. But a mission control room and a high bay, to assemble landers, were complete. The company’s youthful CEO, John Thornton, assured Zurbuchen that everything was on track for the launch of the company’s Peregrine spacecraft in 2022, 1 year later than originally planned.
CLPS, modeled after the commercial cargo and crew programs that pay SpaceX and others to fly to the International Space Station, is Zurbuchen’s brainchild. It pays firms to carry low-cost scientific instruments at fixed prices on their lunar landers, while keeping NASA oversight at a minimum. He proposed the program in 2017, in a canny bid to add to NASA’s science budget while accommodating the incoming Trump administration, which was eager to return astronauts to the Moon.
But there are no guarantees this new model will work. Some scientists complain that the process for choosing instruments and landing sites is opaque, without a traditional peer review from outside scientists. “We want more information on how these sites are selected,” says Amy Fagan, a planetary scientist at Western Carolina University, Cullowhee, and chair of the Lunar Exploration Analysis Group, which advises NASA. She and others also worry the program lacks a long-term plan for its science campaign. “What is the big picture goal?” she asks.
The technical challenges are daunting as well. Engineers and space scientists alike are haunted by 2019 missions to the lunar surface by an Israeli company and the Indian government that ended in dramatic failure, underscoring how difficult it can be to land on the Moon. “We all got scared by the SpaceIL and Indian failures,” says Robert Grimm, a planetary scientist at the Southwest Research Institute, Boulder, who has instruments on two CLPS landers. “There’s smart people all over the world. So what did they do wrong?”
From the start, Zurbuchen had pitched CLPS as risky, saying as many as half of the missions could fail given the limited oversight NASA would have over the landers’ construction and operation. Even so, one or two early crashes could derail the program, triggering a retreat from the White House or Congress. “The No. 1 priority is let’s make sure the first one is successful,” says Clive Neal, a lunar scientist at the University of Notre Dame. For Astrobotic, among the first in line, the pressure is intense.
Thornton got the space itch in 1997, when he was in fifth grade. NASA’s Pathfinder mission had just landed on the surface of Mars, carrying the Sojourner rover. “I thought that was the coolest thing ever,” he says. He built a Lego model of the rover for class.
Later, Thornton studied engineering at Carnegie Mellon University (CMU) in Pittsburgh and worked with William “Red” Whittaker, the famed roboticist. In 2007, as he was nearing graduation, the Google Lunar XPRIZE was announced, promising $30 million in prizes for teams that could land robotic probes on the Moon. Thornton faced a choice: He could go to big aerospace, where he might spend his entire career slowly building a few spacecraft, he says, “or I could take my moonshot.”
He joined Astrobotic, which Whittaker led at the time, as an engineer. In those early days, Astrobotic got by with modest NASA tech development grants. The contracts kept growing larger, building the company’s credibility, and when Thornton took over in 2013 it began to focus on the business of lunar delivery. Companies were delivering cargo to orbit, he says, and he thought, “Why not the Moon?” But it soon became clear that the $20 million top XPRIZE award, with a final deadline of 2018, was far too small to build a business on. “We saw the writing on the wall,” he says.
Ultimately, none of the XPRIZE competitors made it, but Zurbuchen, who was an engineer and space scientist at the University of Michigan, Ann Arbor, before joining NASA in 2016, saw an opportunity. When the incoming Trump administration expressed interest in a return to the Moon, he got on the phone and started asking around to see whether some of the companies that participated in the program were legit. They seemed to be—and if they stayed cheap, he thought, they could give NASA a fast route to the Moon. “It took advantage of that unique opportunity in time,” Zurbuchen says. The administration and Congress were sold.
To keep the effort lean, NASA sent out a call for scientific instruments that were, essentially, sitting on the shelf. Some had been built for a previously proposed lunar rover, Resource Prospector, that was ultimately canceled. Others sprung from unflown instruments designed for sounding rockets, orbiting spacecraft, and even a delayed European-Russian lander. As one NASA official told Grimm early in the program, “We’re so desperate for payloads, we’ll fly rocks back to the Moon.”
In May 2019, NASA selected two small U.S. landers to carry the first of these instruments: Peregrine, from Astrobotic, and Nova-C, from Intuitive Machines in Houston. (A third lander selection was rescinded once it became clear its operations were largely in India.) The two landers would aim for easy, flat, and somewhat boring equatorial sites on the Moon’s near side: Astrobotic to Lacus Mortis and Nova-C to Oceanus Procellarum. Each would be solar-powered, lasting about half a lunar day—roughly 2 weeks on Earth. And they would be cheap, with the government paying less than $100 million, when a NASA-built lander might cost $500 million, or likely more.
The pandemic has complicated life for Astrobotic, like everyone else. Two NASA CLPS contracts allowed the company to grow from 18 people less than 3 years ago to more than 150. But distancing and remote work have slowed development time. “We have a lot of employees that we’ve hired that have never actually been to the office before,” Thornton says.
The delay has also complicated Astrobotic’s hopes to be the first lander: Both it and Intuitive Machines are slated for 2022 landings now, and the latter is launching on a proven SpaceX Falcon 9 rocket, whereas Peregrine has booked the maiden flight of the next-generation Vulcan rocket from the United Launch Alliance, a joint venture of Lockheed Martin and Boeing.
The pace was starting to pick up by the time of Zurbuchen’s June visit. It was a tricky dance for the NASA official. When Zurbuchen visits a typical NASA mission, he likes to reach down deep into the project’s guts to find its top three problems. With CLPS, “We’re not managing that,” he says. “They need to be their own company.” And so he pushes instead for higher level details—how the company is testing interfaces for NASA’s scientific instruments, for example, or how it is interacting with the agency. “Is NASA helping where they’re asking?”
During his visit, a full-scale structural model, which Astrobotic used to test how the lander would handle a rocket launch, sat in the company’s high bay, standing roughly 2 meters tall. Peregrine’s four stubby legs and squat structure are reminiscent of the Apollo landers. It’s an intentionally simple design, based on an aluminum alloy bus with a fixed solar panel on top and no complex moving parts.
Astrobotic once planned to ring the lander with decks that would house its payloads. But the Moon is unforgiving, with no air to carry heat away from sunlit components. To survive on the surface, engineers had to ditch the decks, replacing them with metallic heat shielding and moving around some of the 21 payloads the lander will carry. “It was kind of a push and pull to really find a solution that worked best for us and for them,” says John Walker, a payload manager at Astrobotic.
Adapting to this commercial model has been challenging for scientists. “This isn’t a mission, it’s a bunch of instruments on a lander,” says Barbara Cohen, a planetary scientist at NASA Goddard Space Flight Center who leads development of one of the instruments that will fly on Peregrine, the PROSPECT Ion-Trap Mass Spectrometer. Researchers don’t have a good sense of the order in which the instruments will be turned on and when their measurements will begin. They also can’t dictate the flight path of the lander or which gases, for example, other instruments might use, potentially contaminating their measurements. There’s no project scientist for the overall mission to sort it all out.
At the same time, the loose coordination has allowed Cohen and her colleagues to move fast on completing their instrument, which will hunt for signs of water bouncing around just above the lunar surface. “It’s been a challenge in an interesting way,” Cohen says. “It’s very fast.”
Sometime next year, the spacecraft will ride its Vulcan rocket to lunar orbit, where it will linger for up to 1 month, awaiting the right moment to land—just after the Sun has risen on Lacus Mortis. It will descend flying on its side, then pivot 100 meters above the surface to a legs-down configuration. A laser-ranging system will help guide it safely to the surface, where it will begin its brief, intense life.
If there’s a dominant theme to the early CLPS missions, it’s water. And for good reason. A series of orbital missions over the past 2 decades has detected tantalizing signs of the stuff across the Moon’s surface, defying past notions of a desiccated planetary body. Two NASA missions have turned up evidence of water ice in shadowed areas and near the south pole. And when Chandrayaan-1, India’s first lunar mission, orbited the Moon in 2009, a NASA instrument aboard picked up evidence of water all over the surface, the signal strongest during lunar dawn and dusk.
“I call it space dew,” says Georgiana Kramer, a lunar scientist at the Planetary Science Institute. The Sun throws hydrogen ions at the lunar surface, in the form of the solar wind, that bombard oxygen-rich rocks, apparently forming small amounts of water. The vapor seems to dance across the surface and may gain enough energy to hop from the equator to shadowed cold traps at the poles, although “we don’t really understand its cycle,” Cohen says.
Among the water-focused payloads on Peregrine will be three instruments that will form the bass beat of future CLPS missions, reappearing on various landers and ultimately ending up on NASA’s Volatiles Investigating Polar Exploration Rover (VIPER), which Astrobotic will carry to the Moon as early as 2023. This VIPER suite will map lunar water and probe its origins with a mass spectrometer that can detect water directly; a neutron counter to measure hydrogen based on the neutrons it absorbs; and a near-infrared spectrometer to detect hydrogen in drill tailings.
Peregrine’s mission to Lacus Mortis is unlikely to provide any revelations itself—stuck in one location, it won’t be able to sample multiple areas. It will, however, provide baseline measurements of how much the lander contaminates the surface and prove out how these tools work together. “It’s risk reduction, big time,” for future missions, says Richard Elphic, a space physicist at NASA Ames Research Center and lead on the neutron counter.
Other water-hunting missions will follow close behind. PRIME-1, carried on a lander built by Intuitive Machines, is slated to bring the VIPER mass spectrometer together with a specialized drill to the south pole in November 2022 (see graphic, p. 1191). Another, built by Masten Space Systems in Mojave, California, will launch in 2023. Masten’s lander will use powerful cameras to study the texture and composition of the soil surrounding it and wield a dexterous robotic arm to scoop up samples for analysis. It will also include the three VIPER instruments and deploy a small autonomous rover, built by Astrobotic and CMU, that will carry the neutron counter to scan the first meter of the surface for hydrogen spikes indicating buried water ice.
Soon after, in November 2023, it will be VIPER’s turn. The $430 million NASA rover will be carried by Astrobotic’s Griffin lander, which will be about the size of a sedan. Once it rolls down the ramp, VIPER will be fully independent, running for 100 days and conducting more than 40 drilling operations while constantly counting neutrons. Beyond mapping the sheer amount of water and its distribution—important for future use by astronauts—its mass spectrometer will be able to tease out the water’s origins by looking at hydrogen isotopes. Whereas some is likely created by solar wind, which contains light hydrogen, other water may be delivered by comets, rich in a heavier isotope. And by sniffing for sulfur, the instrument might detect whether any of this water seeped from the Moon’s interior.
VIPER may be the culmination of CLPS’s first stage, but it could also be a turning point. If the program is a success, NASA may turn to companies like Astrobotic not just for transportation to the Moon, but also for rovers. “There’s a good possibility that the VIPER rover is the first and last NASA-built rover for the Moon,” says Anthony Colaprete, the mission’s project scientist at Ames. “And I’m perfectly fine with that.”
Although water may be the most pressing question about the Moon, it is far from the only one. “We have a million and one lunar science questions,” Cohen says. NASA is now developing instrument packages to address them. In 2023, the first lander built by Firefly Aerospace, for example, will carry a heat probe and an electromagnetic sounder to a nearside lava basin called Mare Crisium, which boasts some of the Moon’s thinnest crust. The instruments will provide the first in situ measurements of the interior’s heat since Apollo, which found that the Moon’s surface was surprisingly warm—and help clarify whether the terrain the astronauts explored was an anomaly and how it might have formed.
Instruments to probe the interior are just part of the cargo, leading Brian Walsh, an astronomer at Boston University, to call Firefly a “beautiful Frankenstein.” Walsh’s own instrument, the Lunar Environment Heliospheric X-ray Imager, will look back at Earth and take the first ever picture of the planet’s magnetic field by detecting solar x-rays careening off it. Although Earth’s magnetic field has been measured in many ways, it has never been imaged from the outside. “It’s hard to study from the inside,” Walsh says. “If you were trying to study a whale, you couldn’t do it from its belly.”
Still, coordinated science will be a theme going forward, Zurbuchen says. NASA has already selected two new packages of instruments, which CLPS companies are now bidding to take. The first, called Lunar Vertex, will be carried by a rover across a “lunar swirl.” Even though the Moon lacks an overall magnetic field, these mysterious, dark areas have a localized field that seems to shield them from the solar wind. The second payload, perhaps CLPS’s most scientifically ambitious, will go to the Schrödinger basin, a 316-kilometer-wide impact crater on the Moon’s far side that features a peak ring, a central upwelling of rock that is a hallmark of large impacts. Copies of the heat probe and magnetometer flown on Firefly will land here, teasing out the source of the material in this peak ring—does it come from the crust or the mantle? The answer will illuminate how giant impacts have shaped the surfaces of rocky planets and moons throughout the Solar System.
The Schrödinger mission will also carry the first NASA seismometer to run on the Moon since the 1970s—and the first ever on its far side. It will listen for moonquakes originating from the far side, which Apollo never detected—perhaps indicating deeper geological differences between the two hemispheres, says Mark Panning, a seismologist at NASA’s Jet Propulsion Laboratory who is leading development of the Schrödinger seismometer. The background seismic data will also reveal the rate at which micrometeorites bombard the surface—essential information for shielding long-term lunar inhabitants.
Finally, the lander will carry a small radio telescope. After the lander shuts off and the Sun dips below the lunar horizon, the telescope’s far-side vantage will ensure a quiet unknown on Earth, allowing it to search for cosmological signals of a time before stars, when dark clouds of neutral hydrogen began to slowly pull together. “That’s a very challenging measurement,” says Bale, the instrument’s lead investigator. “I wouldn’t claim a priori that we’ll go to Stockholm. But it’s a goal.”
All that science could come in the next 5 years—a striking contrast to most NASA science missions, which can take decades to develop, launch, and return data. The demographics of lunar scientists reflect that pace. “Go to a lunar science meeting and the faces look very different than at other planetary meetings,” Zurbuchen says. A number of younger researchers are getting involved and writing proposals, drawn by the possibility of getting an instrument to the lunar surface in a matter of a few years. Many haven’t been successful yet, but each round of proposals brings valuable experience, Denevi says. “The opportunity is open.”
How long it will remain open is not clear. The Biden administration supports CLPS. But its focus is not guaranteed to stay on science; among the landers now vying for CLPS contracts are SpaceX’s enormous Starship, a spacecraft that NASA has selected to serve as its lander for astronauts and that might also be used to ferry supplies. Astrobotic also sees a future in providing services for humans on the Moon—delivering permanent solar stations or other equipment. Expanding beyond small science payloads and developing landers that can survive the frigid lunar night will be essential to keeping these companies viable, Neal says. “If you keep CLPS just as small payloads surviving only one [lunar] day, it’s a dead end. It’s going to get cut.”
For now, though, NASA’s focus is on those first launches. At the end of Zurbuchen’s visit, during an evening talk a short distance from Astrobotic’s headquarters, he lauded the company and the innovation it was fostering in Pittsburgh, comparing it to the NASA team that landed Perseverance on Mars. “What happened is an amazing team came together, persevered, and inspired everybody. And John,” he said, looking over at Thornton, “that’s exactly what I expect from you.”
“No pressure,” Thornton said afterward. “I’m sweating already.”