The universe is constantly beaming its history to us. For instance: Information about what happened long, long ago, contained in the long-length radio waves that are ubiquitous throughout the universe, likely hold the details about how the first stars and black holes were formed. There’s a problem, though. Because of our atmosphere and noisy radio signals generated by modern society, we can’t read them from Earth.

That’s why NASA is in the early stages of planning what it would take to build an automated research telescope on the far side of the moon. One of the most ambitious proposals would build the Lunar Crater Radio Telescope, the largest (by a lot) filled-aperture radio telescope dish in the universe. Another duo of projects, called FarSide and FarView, would connect a vast array of antennas—eventually over 100,000, many built on the moon itself and made out of its surface material—to pick up the signals. The projects are all part of NASA’s Institute for Advanced Concepts (NIAC) program, which awards innovators and entrepreneurs with funding to advance radical ideas in hopes of creating breakthrough aerospace concepts. While they are still hypothetical, and years away from reality, the findings from these projects could reshape our cosmological model of the universe.

“With our telescopes on the moon, we can reverse-engineer the radio spectra that we record, and infer for the first time the properties of the very first stars,” said Jack Burns, a cosmologist at the University of Colorado Boulder and the co-investigator and science lead for both FarSide and FarView. “We care about those first stars because we care about our own origins—I mean, where did we come from? Where did the Sun come from? Where did the Earth come from? The Milky Way?”

The answers to those questions come from a dim moment in the universe about 13.7 billion years ago.

When the universe cooled about 400,000 years after the Big Bang, the first atoms, neutral hydrogen, released their photons in a burst of electromagnetic radiation that scientists can still see today. This cosmic microwave background, or CMB, was first detected in 1964. Today scientists use complex tools like the European Space Agency’s Planck probe to detect its minute fluctuations, which create a snapshot view of the distribution of matter and energy in the young universe. Scientists can also fast-forward about a hundred million years to study much of the roughly 13 billion years since the formation of the first stars, or “Cosmic Dawn,” thanks to visual data gleaned from starlight by telescopes like the Hubble (and soon, the upgraded James Webb). They allow us to see so far that we are literally looking into the past.

After the initial fireball from the Big Bang faded into the CMB, but before the first stars started burning, there was a period when almost no light was being emitted in the universe. Scientists refer to this period without visible or infrared light as the “Cosmic Dark Ages.” During this epoch, it seems likely that the universe was very simple, consisting mostly of neutral hydrogen, photons, and dark matter. Evidence about what happened during this period might help us understand how dark matter and dark energy—which by our best guesses make up about 95 percent of the mass of the universe, yet are largely invisible to us and which we still don’t really understand—shaped its formation.

There are clues about what happened during the Cosmic Dark Ages whizzing around, hidden in hydrogen, which still makes up the majority of the known matter in the universe. Each time the spin of a hydrogen’s atom’s electrons flips, it gives off a radio wave at a specific wavelength: 21 centimeters. But those wavelengths released during the Cosmic Dark Ages are not actually 21 centimeters long by the time they reach Earth. Because the universe is rapidly expanding, hydrogen wavelengths also expand, or “red-shift,” stretching out when they travel across vast distances. This means each wave’s length functions like a timestamp: The longer the wave, the older it is. By the time they reach Earth, they are more like ten or even 100 meters long, with frequencies below the FM band.

Despite their low frequency, these waves could be captured by a radio telescope—if our atmosphere wasn’t in the way. The ionosphere, ionized by the sun’s electrical energy, absorbs or reflects this information before it reaches us. Our radio communications on Earth disrupt it, too. So imagine it: From the Dark Ages of the cosmos they travel, ready to tell us what exactly was going on when they were made, and then BLAM—ionosphered. Bye-bye, cosmic truths.

“We are absolutely completely ignorant about the radiation of the universe at long wavelengths that won’t go through our atmosphere,” says John Mather, a cosmologist, astrophysicist, and Nobel Laureate for his work studying the cosmic microwave background. “There could be big surprises out there.”

That’s where the moon comes in. On its far side, it blocks Earth’s radio signals. There is no ionosphere. For incredibly long wavelengths, it’s a perfect port of call.

To capture them, Burns’s FarSide and FarView proposals eschew a solid-aperture radio telescope (imagine the late Arecibo) in favor of a vast array of simple dipole antennas—much like the rabbit ears on your grandpa’s old TV. FarSide would require a 590-kilogram base station and 128 pairs of antennas connected by a tether, which would be unspooled in the shape of four spiral arms across a 10-kilometer swath of the moon. A single lunar rover would handle the construction. The base station would serve as a central processing center for the signal data picked up by the antennas, and would beam it to an alternative relay satellite orbiting the moon.

FarView, a more ambitious program that’s been designed with the help of Houston-based Lunar Resources, Inc., would spread 100,000 dipole antennas across 400 square kilometers of the moon. But the plan isn’t just an upscaled version of FarSide. FarView builds itself—out of the moon. First, a team of automated rovers would gather up regolith and deliver it to a “factory” that could extract aluminum. Another ten or so rovers would fabricate thin antennas out of that metal and then use an electrolysis technique to electroplate them onto the lunar surface. Solar panels to run the system could also be made onsite.

Burns’ idea is to use arrays like these to create a map of specific areas of the universe during the Dark Ages. The longer the neutral hydrogen wavelength, the farther back into time scientists know they’re looking. The wavelengths might also show if the neutral hydrogen that released the wave was warmer or colder than the cosmic microwave background released shortly after the Big Bang; that information might reveal the role dark matter played in the happenings of the Dark Ages, and offer clues about what, exactly, dark matter is. “I like to tell my physics colleagues: ‘Imagine we have just built you a brand new high-energy particle accelerator, and it’s bigger than anything we could ever imagine. Well, the universe did that for us.’ Those particles are there from the Dark Ages and the Cosmic Dawn,” Burns says. “We are going to use our radio telescopes like a particle detector to understand the kind of physics that was operating in this un-sampled time in the universe.”

“This is a very important part of the story of the thermal history of the universe,” agrees Mather. “Was the expansion of the universe cooling this matter, or were objects like stars turning on and warming the matter up again?”

Burns’s twin projects are the endpoint of more than 35 years of research, including an article he wrote for Scientific American in 1990 that laid out the obstacles to building a 10- to 15-meter lunar radio telescope at the time. “I really thought we’d have one of these telescopes on the moon by now,” he says.

But NASA’s push to return to the moon means Burns’s dream may be coming true. So far, both FarSide and FarView have received $125,000 in funding from NASA for initial engineering design studies. In 2022, the agency intends to dispatch a single low-frequency radio spectrometer via a commercial lunar lander. The device is called Radio wave Observations at the Lunar Surface of the photoElectron Sheath (or Rolses), and it will be an important proof of concept for future moon-based radio telescopes. Another radio signal probe called the Dark Ages Polarimeter Pathfinder (Dapper), is proposed as a payload to land on the lunar farside along with the LuSEE radio instrument in 2024. It will capture redshifted 21-centimeter radio wavelengths on the far side before downloading its data to Earth via a lunar-orbiting relay satellite.

But still, there’s an even more jaw-dropping idea: NASA Jet Propulsion Labs’ Lunar Crater Radio Telescope, which just received $500,000 in Round II NIAC funding. It would create the most audacious radio telescope ever built. Its aluminum mesh dish would stretch a kilometer across and 600 meters deep, housed inside a crater 3 kilometers wide. Its parabolic dish would catch long-wavelength radio waves traveling through space and direct them to a receiver suspended over the crater.

Saptarshi Bandyopadhyay, the roboticist who’s the mastermind behind the concept, was inspired by Burns’s 1990 paper on why a radio telescope in a lunar crater wouldn’t work. (Or at least, couldn’t work back then.) Those limitations included finding the perfect crater and the difficulty of constructing the towers required by traditional radio telescope dishes. But finding the right spot is currently being accomplished thanks to the Lunar Reconnaissance Orbiter Mission, and a new design and materials mean towers are no longer required. “Our innovation was saying: ‘Oh look, we can solve all of this now, because we have all these technologies that can take care of these issues,’” Bandyophadhyay says. “If we redesign all of these ideas in this new way, we can make this happen.”

The LCRT would be more expensive and far more complicated to execute than FarSide’s 128-antenna approach. But it would also provide extremely accurate data, giving us a clearer view of, say, how galaxies were formed 12.5 billion years ago. By capturing the longest wavelengths, Mather says, it might map a picture of “a very simple universe, where there were no stars yet, no galaxies, just some blobs” showing the density of dark matter. “Finding that,” he continues, “would be very cool.”

Bandyopadhyay’s team will use the $500,000 to run complex simulations testing different ways rovers might build the enormous dish. They have a pretty good idea of what will work. Instead of a tower, they’ll use a simplified design in which the telescope’s receiver will hang on wires strung across the crater—a spider perched precipitously above its web. The web will be a lattice of aluminum mesh, composed of radial wires running from the lander—situated at the bottom of the crater—up to the rim. Circumferential wires will electrically connect them.

To build it, half of the landing craft, carrying the light, durable mesh that will make up the webbed dish, would land in the crater. The other half, carrying DuAxel rovers designed by JPL, would split off and land at the crater’s rim. The rovers are 4-wheeled workhorses with two axles that can separate and reconnect with each other. Half of each rover would anchor to the rim, then belay its partner down to the main lander on the crater floor. The crawler would attach to the aluminum mesh at the lander, then climb back up the crater, unfurling the web behind it, which could simply unfold, like a giant fishing net. After making its way back up the rim, each rover would anchor the dish’s radial lift wires in place.

And if that won’t work, Bandyopadhyay has a second plan. “Another idea is to not use robots, but to fire harpoons into the crater wall” from the landing craft at the bottom of the crater, he says, with the rovers helping to tension the aluminum mesh dish.

Needless to say, all three project concepts are up against some major challenges. The NIAC funding is just a drop in the bucket; each would cost more than $1 billion to develop, build, and become operational. (“I would like to say to anyone who has money that if you give me $5 billion, I can launch this tomorrow,” Bandyopadhyay says.) 

There’s also the problem of labor. All three projects propose using rovers, which would need to hibernate to survive the -173 Celcius temperatures of the lunar night—which lasts 14 Earth days. And it’s unclear if it would be best to use rovers that are automated, or operated by astronauts on the moon, in orbit, or on Earth. Most of all, the precise strokes of orchestrating not just a successful landing but also a flawless rover-based construction project on a vast scale are … let’s say, yet to be determined.

On an optimal schedule, FarSide could begin operations before the end of the decade; FarView in the 2030s; and the LCRT by 2040. “I would personally be very surprised to see it launch before I retire,” Bandyopadhyay says.

In the meantime, other projects may help us understand the secrets of the Cosmic Dark Ages. The new James Webb telescope, which is expected to launch this fall to study the Cosmic Dawn, may provide data that could help scientists extrapolate backwards into the Dark Ages. And researchers are working to better study the more limited neutral hydrogen frequencies that they can observe from Earth.

But until they either reach the far side or run out of time, Bandyopadhyay, Burns, and others will keep shooting for the moon. “I’m a child of optimism and science fiction,” he says. “I want—not for myself, but for my grandkids or great-grandkids—to enable space travel, matter and antimatter engines, and things like that. And we’ll be nowhere then if we don’t seek answers to fundamental questions like ‘What is the universe made of?’ right now.”