Night by night, star by star, astronomers are edging ever closer to learning just how crowded our universe really is—or at least our galaxy, anyway. A quarter century after the first exoplanets were found orbiting other stars, statistics from the thousands now known have revealed that, on average, each and every stellar denizen of the Milky Way must be accompanied by at least one world. Look long and hard enough for a planet around any given star in our galaxy and you are practically guaranteed to find something sooner or later.

But even a crowded universe can be a lonely place. Our planet-rich Milky Way may prove to be life-poor. Of all the galaxy’s known worlds, only a figurative handful resemble Earth in size and orbit—each occupying a nebulous “Goldilocks” region of just-rightness—a fairy-tale-simple ideal in which a world is neither too big nor too small, neither too hot nor too cold, to sustain liquid water and life on its surface. Instead, most of the Milky Way’s planets are worlds theorists failed to predict and have yet to fit comfortably in any conception of habitability: “super-Earths” bigger than our familiar planet but smaller than Neptune. No super-Earths twirl around our sun for solar system–bound scientists to directly study, making it that much harder to know whether any elsewhere are Goldilocks worlds—or, for that matter, whether any one-size-fits-all metric of habitability is hopelessly naive.

Grappling with these astrobiological mysteries requires new generations of telescopes and spacecraft to seek out and study signs of habitability and life beyond the solar system. But the proof for or against a lonely, crowded universe may be surprisingly close at hand, celestially speaking. In 2016 years of scrutiny at last revealed an Earth-size world right next door, in a temperate orbit around the smallest member of Alpha Centauri, a triple-star system that at 4.4 light-years away is the closest to our sun. Now, another exhaustive search of our solar system’s next-nearest neighbor, Barnard’s Star, just shy of six light-years distant, has uncovered a candidate planet there, too—a bigger, colder super-Earth provisionally dubbed Barnard’s Star b. Achieved by an international team of more than 60 astronomers using observatories from around the world, the planetary discovery is detailed in a study in the November 14 Nature. It opens the floodgates for future investigations of—and comparisons between—the two familiar-but-alien planets closest to our solar system.

A Frozen Super-Earth?

“If you live in a city of millions of people, you are not interested in meeting every one of them—but maybe you want to meet your immediate neighbors,” says lead author Ignasi Ribas, an astronomer at the Institute of Space Studies of Catalonia in Spain. “This is what we are doing for the planetary systems of the stars that surround us. Otherwise we cannot answer the big questions. How does our solar system and our Earth fit in with the rest of the galaxy? Are there other habitable or inhabited planets? Barnard’s Star b is not giving us those answers just yet, but it is telling us part of the story we need to know.”

Located in the constellation of Ophiuchus, Barnard’s Star is so dim in visible light that it cannot be seen with unaided eyes. Yet it has been a favorite of astronomers since 1916, when measurements revealed its apparent motion across the sky was greater than that of any other star relative to our sun—a sign of its extremely close cosmic proximity. The star’s nearness to us is only temporary—within tens of thousands of years, its trajectory will have swept it out of our solar system’s list of top five closest stars.

According to Ribas and his colleagues, the candidate planet is at least three times heavier than our own, and circles its star in a 233-day orbit. That would put it in the torrid orbital vicinity of Venus around our yellow sun, but Barnard’s Star is a comparatively pint-size and dim red dwarf star. This means its newfound companion is near the “snow line,” the boundary beyond which water almost exclusively exists as frozen ice—a region around other stars thought to be chock-full of planets, but that astronomers have only just begun to probe for small worlds.

There, Barnard’s Star b would receive only 2 percent as much starlight as Earth gets from the sun—just enough to give it an estimated average temperature of –150 degrees Celsius. Perhaps, Ribas speculates, the planet is rocky and covered in thick layers of ice, with a surface that resembles those of Jupiter’s and Saturn’s frozen moons. Prospects for life would appear remote on such a world—unless, akin to those selfsame moons, it possesses a subsurface ocean kept liquid by internal heat. In that case, its interior would need to somehow stay warm for a very long time—the planet’s age is thought to be somewhere between six billion and 11 billion years, the general estimates for the age of Barnard’s Star. (By comparison, the Earth is only 4.5 billion years old.)

Alternatively, the planet might be covered by a thick, insulating blanket of hydrogen leftover from its birth in a spinning disk of gas and dust around its star. Although hydrogen on smaller, hotter worlds would dissipate into space, super-Earths in frigid orbits might manage to hang on to enough of the gas to build up a massive planet-warming greenhouse effect—a possibility that throws Earth-centric Goldilocks ideas into tumult. If this mechanism operates on Barnard’s Star b or other cold super-Earths, “our dreams that every star may have a habitable planet could well come true,” says Sara Seager, a planet-hunting astrophysicist at Massachusetts Institute of Technology who was not involved with Ribas’s study. “There are some crazy worlds out there.”

Haunted by History

Some worlds, alas, are simply too good to be true. In 1963, the Dutch astronomer Peter van de Kamp famously “discovered” planets around Barnard’s Star—linking supposed shifts in the star’s motion in the plane of the sky to the gravitational influence of unseen worlds. By the 1970s, the evidence for van de Kamp’s putative planets was evaporating under further scrutiny, in the end being attributed to various errors and oversights that contaminated his observations. Through it all, van de Kamp’s faith in his claims was seemingly unshakeable; he continued to insist the planets were genuine for the remaining decades of his life.

That cautionary tale still haunts today’s planet hunters. Fears linger, even though modern-day evidence for a world around Barnard’s Star is far more robust. After all, if history does not repeat itself, it can all too easily rhyme. “In light of the troubled history of planet claims for this star, the authors are very brave to stick their necks out like this,” says Ignas Snellen, an astronomer at Leiden University in the Netherlands who did not take part in the research. “These are very difficult measurements!”

So difficult, in fact, that some experts remain unconvinced. “Since there are planets everywhere, I suppose that there must be planets around Barnard’s Star,” says Debra Fischer, an astronomer and veteran planet hunter at Yale University who was not involved with the putative discovery. “There may even be one a few times the Earth’s mass with a period of 233 days. But this analysis doesn’t provide strong enough support, in my opinion.”

In contrast, Xavier Dumusque, an astrophysicist at Geneva Observatory in Switzerland also unaffiliated with Ribas’s study, finds the evidence for Barnard’s Star b compelling. “In terms of likelihood that this planet exists, I think there can be no doubt,” he says. “Its signal is really clear.”

The case for Barnard’s Star b hinges on a remarkable feat of data collection and analysis that stitches together hundreds of measurements from seven world-class instruments on large ground-based telescopes across a span of more than 20 years. Each measurement tracks the radial velocity of Barnard’s Star—its motion toward or away from Earth, which can wobble back and forth in sync with the orbital tugging of accompanying planets. The signal attributed to Barnard’s Star b is a wobble of just over a meter per second—a walking-speed-scale effect readily mimicked by various kinds of stellar activity or instrumental errors. Its appearance across two decades of data from so many different surveys strongly suggests the signal is not due to instrumental noise, but definitively ruling out stellar activity as its source is much harder. Even the most elite radial-velocity planet-hunting teams have been stung by such star-induced false positives in the not-too-distant past, claiming sensational new worlds that ultimately proved illusory.

Here, Barnard’s Star may offer an advantage despite its star-crossed planet-hunting history. It is actually one of the most quiescent stars known, making it nearly ideal for radial-velocity work. And Ribas and colleagues insist they have learned the hard and necessary lessons from past claims of phantom worlds. An intensive array of follow-up observations has largely ruled out the influence of starspots and other obvious sources of planetary mimicry, Ribas says, and the study’s authors also conducted over half a million simulations to conclude the signal’s chance of arising from more pernicious stellar effects is less than 1 percent. “I am over 99 percent certain that the planet is there,” Ribas says. “But we do have the story of Peter van de Kamp close in our minds. If someone makes strong arguments against our discovery, I will concede! I would not like to become the van de Kamp of the 21st century.”

Time to Take a Picture?

One way or another, certainty about this controversial planetary candidate could come soon. Already, the team’s work has ruled out any Earth-size planets in orbits of 40 days or less around Barnard’s Star, although it also found wobbly, as-yet-unconfirmed hints of another planet lurking much farther out. (With apologies to van de Kamp, such a planet would probably still be a poor match for his prior claims.) Hundreds of additional radial velocity measurements with existing and upcoming instruments could further increase confidence in the candidate’s reality, as could forthcoming data from the European Space Agency’s Gaia spacecraft, which tracks the motions of Barnard’s Star and more than a billion other stars in its efforts to make a three-dimensional map of the Milky Way.

Although exceedingly unlikely, the planet could by chance be precisely aligned with our perspective from Earth so that it would “transit”—casting a detectable planetary shadow toward our telescopes as it flits across the face of Barnard’s Star. Demonstrated to stunning success by NASA’s late Kepler spacecraft and hailed as a key enabler of future exoplanet observations with the space agency’s Transiting Exoplanet Survey Satellite and the James Webb Space Telescope, the transit technique has delivered most of the worlds in astronomers’ catalogues. But most planets do not transit as seen from Earth, particularly those that are in wide orbits around their stars like Barnard’s Star b.

The planet’s relatively wide separation from its star, however, offers an even more promising and tantalizing possibility—the prospect of taking its picture, or “direct imaging” as an astronomer might say. A snapshot of Barnard’s Star b could reveal many otherwise-inaccessible things, most importantly its true nature—a frozen super-Earth, a hydrogen-swaddled greenhouse world or perhaps something theorists have yet to even dream of. With such an image, astronomers could come one major step closer to solving the mystery of our crowded universe’s loneliness.

In the 2020s a new generation of extremely large ground-based telescopes will come online that could be up to the task. Each will be equipped with a starlight-gathering mirror some 30 meters or more in diameter that could distinguish the planet’s faint photonic emissions. Except, experts say, the first instruments taking planetary pictures on such observatories will be optimized for thermal imaging—a poor choice for seeking out a probably-icy world. Instead, the best hopes may rest with NASA’s next planned post-Webb space observatory, a sort of supercharged Hubble Space Telescope dubbed WFIRST.

That is, if it launches at all—the White House has attempted to cancel the project in recent budgets. The current plan calls for WFIRST to include a coronagraph, a starlight-blocking instrument to allow the light from a select few exoplanets to be seen—all relatively humdrum giant worlds, due to the lack of any suitable smaller candidates existing around nearby stars. But if Barnard’s Star b is real, imaging it with WFIRST “might just be feasible,” says Jeremy Kasdin, a Princeton University astrophysicist leading WFIRST’s coronagraph development. “Everything has to be just right… But in the best-case scenario, it would be challenging but possible.”