Read our COVID-19 research and news.

This image reveals the black hole at the center of Messier 87, a massive galaxy in the nearby Virgo galaxy cluster. The black hole resides 55 million light-years from Earth and has a mass 6.5 billion times that of the sun.

Event Horizon Telescope collaboration et al.

For the first time, you can see what a black hole looks like

At last, we can see it: a black hole in the flesh. Astronomers today revealed a picture of the gargantuan black hole at the heart of the nearby galaxy Messier 87 (M87). The result—a ring of fire surrounding the blackest of shadows—is a powerful confirmation of Albert Einstein’s theory of gravity, or general relativity, which was used to predict black holes 80 years ago. It is also a feat for the team of more than 200 scientists who toiled for years to produce the image by combining signals from eight separate radio observatories spanning the globe.

“It feels like looking at the gates of hell,” says Heino Falcke of Radboud University in Nijmegen, the Netherlands, one of the leaders of the Event Horizon Telescope (EHT) collaboration, which announced the result in a global set of coordinated press conferences. “This is the end of space and time.” Falcke says the 2-year process of crunching the data and generating the images “was the most emotionally difficult period of my life.”

Although few doubted the existence of black holes, seeing them—or at least their shadow—was an immense challenge. Black holes have gravitational fields so strong that even light cannot escape, so they are defined by the shell of a black, featureless sphere called an event horizon. But the holes can nevertheless be seen. As they consume matter that strays too close, they squeeze it into a superheated disk of glowing gas.

In the team’s images, the bottom of the ring appears bright because the gases there are being Doppler-boosted, whipped toward Earth. The black hole bends light around it, creating a circular shadow. General relativity predicts that the shadow ought to be round to within 10%, says Avery Broderick, an EHT member and astrophysicist at the University of Waterloo in Canada, whereas alternative theories of gravity predict distorted, noncircular shapes. The observed shadow is essentially circular, Broderick says.

Data from the South Pole Telescope, one of the radio dishes used in the Event Horizon Telescope, overwintered in Antarctica before being combined with other data.

Junhan Kim/University of Arizona

The EHT team, from 13 institutions around the world, made its observations of M87* and the black hole at the center of our Milky Way, known as Sagittarius A* (Sgr A*), over 5 nights in April 2017 using eight radio telescopes that are sensitive to wavelengths of about a millimeter. At that specific radio frequency, radiation can penetrate the haze of dust and gas that surrounds the centers of galaxies.

But zooming in on the black holes was still a challenge. Black holes pack an immense amount of mass into a surprisingly small space. The black hole at the center of M87, 55 million light-years away, has swallowed the mass of 6.5 billion suns. Yet its event horizon is only 40 billion kilometers across—about four times the diameter of Neptune’s orbit.

No existing telescope has the resolution to see such a distant, tiny object. So, the EHT team coopted most of the millimeter-wave telescopes worldwide and combined their data to produce a virtual telescope the size of Earth through a process called very-long-baseline interferometry. The telescopes they used stretched from Hawaii to Arizona, Mexico to Spain, and Chile to the South Pole. “You can think of them as silvered spots on a global mirror,” says Shep Doeleman, the EHT’s project leader at the Harvard-Smithsonian Center for Astrophysics in Cambridge, Massachusetts. “Then the Earth turns so we can fill in the image.”

The collaboration had made earlier observations with fewer telescopes, but 2017 was the first time they had a globe-spanning array that included the power of the Atacama Large Millimeter/submillimeter Array in Chile with its 64 dishes. Millimeter waves are affected by clouds, so getting good weather was important. In April 2017, the weather gods smiled. “It was one of the smoothest parts of the project,” says team member Feryal Özel of the University of Arizona in Tucson. “Some crews worked 16- or 18-hour shifts, but the whole thing was lucky,” she says, adding: “Analyzing the data was much harder.”

That process has taken the whole of the time since. The volume of data was so great that it could not be transmitted to large computers at the Massachusetts Institute of Technology’s Haystack Observatory in Westford and the Max Planck Institute for Radio Astronomy in Bonn, Germany. Instead, it had to be recorded on disk and shipped, which posed a problem for the South Pole Telescope. It was in lockdown for the austral winter so researchers didn’t get their hands on its data until almost the end of 2017. A total of 4 petabytes were recorded, each reading time-stamped using an atomic clock. If those data were music recorded as MP3s, they would take 8000 years to play.

Photon ring Shadow edge Event horizon Simulations (bottom) helped connect the EHT’s fuzzy image (middle) to a physical model of M87’s black hole (top), and suggest that the accretion disk spins clockwise. Connecting the dots Strange beast Receding jet Jet rotation Approaching jet Accretion disk Supermassive black hole Disk rotation Material rotating toward Earth is Doppler boosted and brighter. Size of the heliopause, the edge of the solar system The Event Horizon Telescope (EHT) team took 2 years to produce an image of the black hole at the center of nearby galaxy Messier 87 (M87), which feeds on a swirling disk of bright matter. Its gravity is so strong that photons orbit it, creating a bright ring. Gravitational lensing magni­fies the black hole’s event horizon into a larger dark shadow, which may be partially ­filled by material in front of the hole.

“It was a pretty gruesome process to crunch all the data,” Falcke says. Powerful processors called correlators compare readings between pairs of telescopes at different distances and orientations to the black holes. Özel compares it to building up a 3D image of the body with a computerized tomography scan, but in this case they do not have all the orientations they need. “We had to make sure we were not filling in the data in a way that could influence interpretation,” she says. Monika Mościbrodzka, the EHT working group coordinator at Radboud University, says four independent teams duplicated the data processing to eliminate biases. She says the result was convincing because, over 4 days of observations of M87*, the shape and size of the shadow was consistent, and the contrast between the bright ring and dark shadow was as large as theory predicted.

The team did not report results for our galaxy’s giant, Sgr A*. Although it is much closer than M87*, it is about 1000 times less massive, with a smaller event horizon. Moreover, it moves more quickly across the sky, complicating observations. Doeleman says the team will turn to Sgr A* next. “We’re not promising anything,” he says. “But we hope to get to it soon.”

Einstein disliked the idea of black holes. Months after he published his theory of general relativity in 1915, German physicist Karl Schwarzschild came up with a solution for Einstein’s equations that suggested that within a certain distance of an infinitesimal point of mass, gravity should be so strong it would stop anything from escaping, even light.

However, for decades, most physicists and astronomers thought such an idea was just a mathematical curiosity. It wasn’t until 1939 that U.S. physicist J. Robert Oppenheimer and colleagues predicted that a massive star could actually collapse to a point.

The idea got a shot in the arm with Jocelyn Bell Burnell’s 1967 discovery of pulsars—dense, spinning neutron stars—which proved the existence of extremely dense, compact objects. Since then, astronomers have accumulated plenty of indirect evidence for the existence of black holes, from the effects of their gravity. Astronomers have found binary systems, such as Cygnus X-1, where a star orbits an unseen, denser object that appears to be gorging itself on material from its stellar partner.

More evidence came from studies of Sgr A*. Over the past couple of decades, observations of a handful of stars in tight, fast orbits leave little room for anything other than a supermassive black hole at the galactic center, one with a mass of about 4 million times that of our sun.

The most compelling evidence came in 2015, with the detection by the Laser Interferometer Gravitational-Wave Observatory of ripples in space-time emitted by the cataclysmic merger of two black holes. With today’s announcement, however, astronomers finally have visual evidence. “I’ve always wanted to see that damned thing,” Falcke says.

Future EHT observations could shed additional light on the nature of black holes. The team hopes to measure the spin and magnetic polarization of the black holes. At M87*, a more voracious and active black hole than Sgr A*, the team could learn about the mechanism that accelerates jets of material out from the poles of the black hole, like beams from a light house. Sera Markoff, an EHT team member and theoretical astrophysicist at the University of Amsterdam, notes that M87* is also an “active galactic nucleus” whose luminosity waxes and wanes as it slurps up matter. “We just got lucky,” she says. “If it had been flaring we might have seen something very different and it may have blocked the shadow.”

The team’s campaign in 2018 was mostly a washout because of bad weather. This year, observations were abandoned because several telescopes were not operating. But next year’s observations should include new telescopes, and they will also begin to observe at shorter wavelengths, which should offer sharper images, Doeleman says. “We’ll be able to extend that image of that shadow out to where it connects to that jet.”

Astronomers outside the EHT team will be eager for unexpected discoveries that could point to theoretical breakthroughs. When asked about the team’s results, Avi Loeb, director of the Black Hole Initiative at Harvard University, says he is most surprised by the lack of surprises. A decade ago, he helped simulate M87*, and he says his images looked much like the EHT’s today. Even so, he says, the team’s result is an important milestone. “An image is worth a thousand words, and seeing is believing,” he says. “Now, we’ve nailed the map of a black hole.”

With additional reporting by Adrian Cho and Dennis Normile.