You won’t hear them say it, but some of the world’s most acclaimed astronomers have been frustrated for the better part of two decades. In that time they and their colleagues have found thousands of exoplanets—planets orbiting stars other than our sun—and have statistically surmised that hundreds of billions more await discovery in our galaxy alone. Swat any star hard enough with state-of-the-art instrumentation, it seems, and it will eventually spill out new worlds like candy pouring from a split piñata. The planet hunters have already unwrapped some of their haul, using telescopes to closely study and even image a small number of unappetizingly uninhabitable gas-giant worlds. Yet so far they have failed to sample the sweetest, most tantalizing pieces to fall in their laps—a handful of potentially Earth-like worlds that could harbor life.

Not that they haven’t tried. Even if only composed of a single noisy pixel, a picture snapped of a promising planet around another star would go a long way toward telling whether that world is really habitable, or even potentially inhabited. It could be the first glimmer of the greatest discovery in human history—proof that we are not cosmically alone. Alas, today’s best telescopes have fallen short. Their large and sophisticated optics are still too small and simplistic to distinguish the faint form of a rocky world whirling amid the glare of a star. Something bigger and bolder seems to be required. To find another Earth, the thinking goes, one must first build a planet-imaging telescope of such size, sophistication and cost that it becomes too big to fail. To build such a telescope, however, one must first find another Earth it could conceivably study. If “show us the money” is the planet hunters’ plea, “show us the planet” is the surefire policymakers’ reply. This catch-22 has left astronomers planning how they’ll image and study Earth-like planets in a future that seems destined to never arrive.

No longer. Due to a single discovery announced last month, that future may play out over the next several years with astronomers using existing and under-construction telescopes on the ground and in space, rather than in fanciful far-future observatories. The catalyst for this epochal transition is Proxima b, a newfound small planet orbiting Proxima Centauri, which at just over four light-years away is the star nearest to our solar system. “Proxima b has drawn a chalk line high up on the wall, and we are all now jumping to reach it,” says Matthew Kenworthy, an astronomer who works on new planet-imaging techniques at Leiden Observatory in the Netherlands. “E-mails are flying back and forth, people are dusting off methods they came up with years ago for seeing a rocky planet that were shelved for lack of finding any nearby. You can practically hear the frenzied brain waves passing through the air. It’s energizing for everyone to know there’s this beautiful, very exciting target right next door.”

For now, astronomers have yet to actually see this new planet—instead, they have simply measured how its to and fro orbital tugging causes Proxima Centauri to wobble back and forth in the sky. That wobble is a whisper that speaks volumes, revealing that this world is just one third more massive than Earth and resides in an 11-day orbit some seven million kilometers from its star. Earth would be scorched if it were so close to the sun but Proxima Centauri is a much smaller, dimmer bulb—a red dwarf star, the most abundant variety in the Milky Way. So Proxima b’s 11-day year exposes it to two thirds as much starlight as Earth—enough to place the planet in the middle of its star’s “habitable zone,” a temperate circumstellar region where liquid water and life could conceivably exist on a rocky world’s surface.

All of which means that Proxima b is far more than the nearest neighboring planet outside our solar system. It is also the premier example of what may be the most common habitat for life in the universe and the closest possibly Earth-like world we can see. The present rush to study it is a sign of things to come, a preview of the 2020s and 2030s when astronomers will strive to image tens if not hundreds of other promising worlds around nearby stars. “This is only one planet, and we don’t yet know whether it is actually habitable or not, but it still is an extremely big deal because it will rapidly push the field into new frontiers,” says Olivier Guyon, a planet-hunting astronomer at the University of Arizona. “We may find other candidates perhaps even better than this one, but for now Proxima b offers a best-case scenario for imaging and studying a potentially Earth-like planet within the next decade.”


The best-case scenario would be if Proxima b “transits,” periodically flitting across the face of its star to cast a shadow toward us across the light-years. Using that planetary silhouette, astronomers could then pin down Proxima b’s exact size and mass, and thus its density—and thereby learn whether it is a ball of rock, a gas-shrouded orb or something in between. They could even determine the bulk chemical composition of Proxima b’s upper atmosphere.

Glimpsing the airy annulus around a transiting Proxima b requires the light-gathering power of what astronomers sometimes call “big glass”—a gigantic mirror such as the 6.5-meter aperture of NASA’s James Webb Space Telescope, set for launch in 2018. But astronomers cannot see what is not there, and the planet is far from guaranteed to possess an atmosphere at all. Proxima b’s star-hugging orbit may give it a surface temperature amenable for water, but also subjects it to atmosphere-eroding stellar flares and tidal forces that lock the planet’s rotation, leaving the side always facing the star airless and eternally bathed in starlight while the atmosphere freezes out on the world’s cold, permanently dark side. Furthermore, chances are slim that Proxima b transits at all—to see the planet’s shadow we would have to view its orbit essentially edge-on, like watching the rim of a spinning record on a light-years-distant turntable.

Unfortunately, a new study released today led by astronomer David Kipping of Columbia University has all but ruled out the possibility of transits for Proxima b. Kipping and his colleagues went searching for its transits in data gathered by Canada’s MOST space telescope, and even found what seemed to be a compelling signal—but additional data from the ground-based HATSouth telescope array suggested the signal was due to Proxima Centauri’s flares rather than any transiting world. “Although we found a candidate transit in the MOST data, HATSouth data rules it out at a confidence level of 70 to 90 percent,” Kipping says. “I would now give it less than a 1 percent chance of being a transiting planet. That sounds very small, but scientists prefer to deal with numbers like 0.0001 percent when we talk about ruling things out, so a more conclusive answer will probably require more data.” That could come soon, if astronomers use NASA’s Spitzer Space Telescope to learn once and for all whether Proxima b transits its star.

In the meantime Avi Loeb and Laura Kreidberg, two astrophysicists at the Harvard–Smithsonian Center for Astrophysics, have already detailed a near-term plan for studying Proxima b without a transit. In a new paper they describe how NASA’s infrared Webb telescope could soon determine whether Proxima b has an atmosphere—an observation also proposed by Proxima b's discoverers and collaborators in an earlier paper. The telescope could not directly see the planet but rather spy some of its infrared light shining through the nearby star’s glare. Assuming Proxima b is tidally locked, Webb could then detect changes in the planet’s thermal glow as its cold, nightside and warm, sunlit dayside shift in and out of view across one complete orbit, rather like watching phases of the moon as it circles Earth. A small temperature difference between the two sides of Proxima b would suggest the presence of an atmosphere or ocean to redistribute heat whereas a large thermal contrast would indicate the planet is a dry, airless rock. “You could potentially see this after looking for only 11 days,” Kreidberg says. “It’s a short and sweet observation to make.”

If Proxima b proves to have an atmosphere, Loeb and Kreidberg have also proposed using Webb to probe for the infrared signature of ozone in Proxima Centauri’s glare as a possible sign that the planet’s air is filled with oxygen—something that, on Earth, is mostly produced by life. But instead of requiring 11 days of observations, Webb’s search for ozone would consume an estimated 60 days—an enormous and risky investment of observing time. “We’d have to use all of Webb’s power to make a measurement like that—everything would have to go just right,” Kreidberg says. “For any other star, this would be a crazy idea but because Proxima Centauri is our nearest neighbor we have a shot. In the big scheme of things two months of observing time on our best space telescope might be worthwhile if it reveals something there associated with life.”

Loeb also has more visionary (and longer-term) plans for Proxima b: send a probe. The planet is an “obvious target” for Breakthrough Starshot, a private venture funded by the billionaire Yuri Milner that seeks to launch fleets of miniature interstellar spacecraft. Loeb serves as chairman of the Starshot advisory board. “You might think that building a big telescope on the Earth could be substitute for sending a camera to a planet,” Loeb says. “But past experience with NASA’s planetary missions tells us this isn’t true—we always learn something new from visiting a place directly. And Starshot will aim to do that.”

According to Mark Clampin, a Webb project scientist and director of astrophysics at NASA Goddard Space Flight Center, Breakthrough Starshot might have better odds than any search for ozone on Proxima b using Webb. “The excitement about looking at a planet in the habitable zone of the star nearest to us gets people geared up,” he says. “But there are a lot of competing science priorities for Webb, and peer reviewers will ask whether Proxima b is really the best object to investigate like this using so much time.” By the time Webb is operational, Clampin says, another NASA mission, the Transiting Exoplanet Survey Satellite (TESS), slated for launch in 2017, will already be producing a short list of other potentially habitable rocky planets around nearby small stars. TESS will mostly search for planets elsewhere in the sky away from Proxima Centauri, near the ecliptic poles, regions directly above and below our solar system that are easy to continuously monitor with most space telescopes. A rocky planet transiting a nearby star there could be far easier to investigate, pushing Proxima b lower on the priority list of Webb’s mission planners.


As useful as Webb might be for studying Proxima b, most astronomers are far more optimistic about using a coming generation of ground-based Extremely Large Telescopes (ELTs), behemoths with mirrors up to 40 meters wide, scheduled to debut in the mid-2020s. Three ELTs are under construction but only two are in the Southern Hemisphere where Proxima Centauri can be seen, specifically in Chile. One, the Giant Magellan Telescope (GMT), is being built by a consortium of U.S., Asian and South American institutions and will use seven 8.4-meter bowl-shaped mirrors forming a light-gathering surface 24.5 meters wide. The other, the European Extremely Large Telescope (E–ELT), is being built by the European Southern Observatory and will link nearly 800 hexagonal 1.4-meter segments to form a 39-meter mirror—the world’s largest. “It’s a good thing that we’re taking different technological approaches,” says Patrick McCarthy, an astronomer at the Carnegie Observatory and interim president of the Giant Magellan Telescope Organization. “If you only see this planet with one of the telescopes, you’ll scratch your head and wonder if it’s an artifact [of instrumental error]; if you see it in both, you’ll know it’s real.”

Trapped beneath Earth’s starlight-blurring atmosphere, neither ELT could naturally achieve the sharp, stable images required to directly glimpse Proxima b—which would appear to be just 38 thousandths of an arc second from its star (an arc second is one 3,600th of a degree). That both telescopes could conceivably glimpse Proxima b at all is due to the planet’s proximity as well as adaptive optics—computer-controlled deformable mirrors that change their shape 1,000 or 2,000 times a second to correct in real time for the turbulent air. Yet the planet would still be more than 10 million times dimmer than its nearby star. Equipped with a coronagraph—a device that blocks starlight similar to the way a thumb held in front of your eye can blot out the sun—the GMT or the E–ELT could strip out the glare of Proxima Centauri to gather the faint photons from Proxima b, forming a dotlike image of the planet for all to see. The dot’s color could be crudely diagnostic, with one hue suggesting a global ocean whereas another might portend vegetation-covered continents or arid plains of sunbaked rock. Over time they could gather enough different-colored photons from the dot to build a spectrum, looking in its shifting hues for signs of water vapor, carbon dioxide, oxygen and other vital gases—signs, that is, of whether or not that world could harbor life as we know it.

Unfortunately, all coronagraphs still leak some starlight, sending unwanted photons playing across mirrors to pile up around edges and imperfections—and forming diffractive speckles that can mask or mimic a real planet. How easily astronomers can distinguish and study Proxima b’s planetary dot from the resulting swarm of phantoms will depend a great deal on the fine design details of each telescope’s mirrors, coronagraphs and instruments.

Many of those details remain in flux, in part due to the discovery of Proxima b and other less-sensational worlds around red dwarfs. To date, most coronagraph development has focused on imaging worlds around bright sunlike stars, where the star–planet contrast is far higher but offset by wider star–planet separations. Imaging red-dwarf worlds like Proxima b calls for different coronagraph designs scarcely sprung from their laboratory cradles, ones with less-demanding contrast levels but more extreme acuity to catch planets orbiting practically cheek to cheek with their dim, cool stars. This is in part why the GMT has yet to settle on a design for a planet-imaging camera whereas the E–ELT’s premiere offering, EPICS (for Exoplanet Imaging Camera and Spectrograph), technically remains only a candidate instrument awaiting formal approval from the European Southern Observatory (ESO).

According to EPICS team leader and ESO astronomer Markus Kasper, the instrument’s eight-to-10-year construction period is scheduled to start in 2019, meaning its observations of Proxima b would only begin in the late 2020s. Even if the instrument were ready sooner, its telescope might not be: The E–ELT is set to begin operations no earlier than 2025, placing an image of Proxima b at least a decade if not 20 years into the future.


For some, that is too long to wait. Christophe Lovis, an astronomer at the University of Geneva, believes there is a way to observe Proxima b from the ground years sooner—around the same time the Webb telescope could study it from space—using current telescopes much smaller than the planned ELTs.

Imaging Proxima b with a ground-based telescope several times smaller than an ELT is only conceivable due to a fundamental quirk of optical physics: To produce an image of any given resolution, shorter wavelengths of light such as those visible to our eyes require smaller mirrors and other optical components than longer wavelengths do. This means a modern eight-meter telescope working in visible light could theoretically see close enough to Proxima Centauri to pluck its planet out of the stellar glare. Unfortunately, observing exoplanets in visible light rather than in longer wavelengths also requires better coronagraphs as well as much faster and more accurate adaptive-optics systems. Visible light’s extreme requirements for adaptive optics are why the ELTs will initially observe most exoplanets in near-infrared light—visible-light adaptive optics systems for their giant mirrors remain decades away.

The best planet-imaging project on Earth is now at ESO’s Very Large Telescope (VLT) in the high desert of northern Chile, where an instrument called SPHERE uses adaptive optics and coronagraphs to snap near-infrared and visible-light pictures of bright, young, giant exoplanets glowing red-hot from their recent formation. SPHERE presently falls short of seeing Proxima b by about a factor of 10 in star–planet contrast and a factor of six in star–planet separation but Lovis has a breathtakingly plausible plan for bringing the instrument up to snuff. Its crux: pair SPHERE with another item slated to debut next year on the VLT—a high-resolution spectrograph called ESPRESSO, for which Lovis is the instrument scientist. “We don’t have to wait 10 or 15 years for the ELTs. We can make feasible, realistic upgrades to SPHERE and ESPRESSO at the VLT and begin studying Proxima b within probably a few years,” Lovis says. “It would be foolish not to make use of these existing instruments, because together they probably offer the fastest and most cost-effective way to get images of this planet.”

Unlike SPHERE, ESPRESSO does not take pictures. It rather looks for planets via wobbling stars—the same technique used to detect Proxima b in the first place. Such instruments don’t actually watch stars jiggle in the sky; instead they precisely measure the color of the star’s light, which becomes bluer or redder as a planet tugs its star closer to or farther from Earth, similar to the Doppler shift that changes the pitch of an ambulance’s siren as it approaches and speeds past. Fed with SPHERE’s observations of Proxima Centauri, ESPRESSO could search for the reflected visible light from Proxima b, which due to the planet’s orbital motion would exhibit a roughly 50 kilometers-per-second Doppler shift from the leftover starlight. Finding and isolating that signal would guide astronomers as they scrub away the rest of the residual starlight and contaminating effects of Earth’s atmosphere. The end result would be a clean, clear image of the planet that SPHERE alone could never produce.

Working with co-authors including Ignas Snellen, an astronomer at the University of Leiden who first demonstrated this technique on giant planets in 2014, Lovis has detailed his plan in a new paper, calling for upgrades to SPHERE’s adaptive optics as well as for multiple fiber-optic connections between the two instruments. Given a few months total of telescope time stretched across perhaps three years, Lovis and Snellen say, they could image Proxima b and probe the planet’s atmosphere for signs of oxygen, water vapor and methane—all crucial measurements for determining whether that faraway world is actually much like Earth at all. Even if their proposal fails to image Proxima b, Lovis and Snellen say it will still benefit planning for the future operations of the E–ELT. Other European astronomers are pursuing their own plans, examining whether planned upgrades to a different VLT instrument called CRIRES or even a small Proxima-dedicated observatory could offer an alternate path to images of the planet.

Some of the same institutions behind the GMT, the U.S.-dominated competitor to ESO’s E-ELT, are also considering a push to image Proxima b using an even smaller ground-based telescope than the VLT. Jared Males, an astronomer at the University of Arizona, plans to look for the planet in visible light with one of the twin 6.5-meter Magellan telescopes in Chile. In attacking the air-blurred starlight, his weapon of choice will be a run-of-the-mill adaptive optics system augmented by another more “extreme” version that uses 2,000 computer-controlled actuators to correct atmospheric distortions by flexing a deformable mirror more than 3,600 times per second. No other adaptive optics system is faster but even this might not be sufficiently speedy to push far enough into the visible spectrum to allow Magellan’s relatively meager mirror to gather Proxima b’s light.

“I never actually thought we could image an Earth-sized planet around Proxima Centauri, but then again I never knew until recently there was an Earth-sized planet there,” Males says. “Now I just think it will be very, very hard—pretty much at the theoretical limit of what is possible, if everything goes perfectly. No one is saying this is a slam dunk case. But characterizing a terrestrial planet in the habitable zone of our sun’s nearest neighboring star will be one of the most important advances in the history of science. This is how we get ready to do it.”