Now, thanks to help from machine-learning technology, they’ve released an updated version of that image that’s much crisper—and, potentially, more useful for unraveling the mysteries of these unusual celestial objects. Researchers published the new image Thursday in The Astrophysical Journal Letters.
The black hole in question is situated roughly 54 million light-years away from Earth, at the center of the Messier 87, or M87, galaxy. The colossal black hole is massive—with roughly 6.5 billion times more mass than the sun—and has a powerful gravitational pull that nothing can escape.
Astronomers have been studying the M87 black hole for several years using the Event Horizon Telescope, a collaborative initiative that brings together data from several ground-based radio observatories located around the world. When astronomers combined the observations from these different sites, they essentially created a powerful, Earth-sized super telescope.
To generate the 2019 image, astronomers used a “generic” machine-learning system to make sense of Event Horizon Telescope data, as Meghan Bartels writes for Scientific American. That process was a good start at visualizing the black hole, but astronomers couldn’t glean much information about it from the blurry result.
They decided to see if they could improve upon that initial view. To create the new image, astronomers used the same data as they did back in 2019, but they fed it into a sophisticated new image reconstruction algorithm.
This new method—called principal-component interferometric modeling, or PRIMO—produced a much crisper image of the “black hole shadow,” or the radiation released by matter falling into the black hole, as study co-author Lia Medeiros, an astrophysicist at the Institute for Advanced Study, tells the Washington Post’s Joel Achenbach. Since it’s impossible to see black holes themselves, this new view helps further strengthen the theory that a black hole is, indeed, present in M87.
Visually, the main difference between the two images is that, in the updated view, the fiery ring is much thinner.
“I affectionately refer to the previous image as the ‘fuzzy orange donut’ and have been referring to this image as the ‘skinny donut,’ which sounds incredibly unappetizing,” says Medeiros to Reuters’ Will Dunham.
PRIMO works by analyzing huge amounts of training data via a process known as dictionary learning, according to the Institute for Advanced Study. More specifically, scientists fed computers 30,000 simulated images of black holes sucking up surrounding gas particles to help PRIMO identify patterns and rules around how black holes draw in matter.
Essentially, PRIMO “provides a way to compensate for the missing information about the object being observed, which is required to generate the image that would have been seen using a single gigantic radio telescope the size of the Earth,” says Tod Lauer, an astronomer at the National Science Foundation’s National Optical-Infrared Astronomy Research Laboratory, or NOIRLab, in a statement.
The new image is a triumph in its own right, scientists say, but it should also help lead to further scientific advancements. As Dennis Overbye reports for the New York Times, scientists are now studying the image to get a more accurate estimate of the black hole’s mass, for instance.
PRIMO will likely also help astronomers get a sharper picture of other black holes, including Sagittarius A*, which is located at the center of our Milky Way galaxy. Scientists released the first image of that black hole last spring.
Still, PRIMO is far from perfect. Some researchers worry that while it’s producing crisper images, the technology may not necessarily be making the view more accurate, per the Washington Post.
Further testing—and more data—could help give astronomers more confidence in PRIMO. But in the meantime, it’s “potentially a very exciting result,” as Ziri Younsi, an astrophysicist at University College London who works on the Event Horizon Telescope but who was not involved in developing the new image, tells Scientific American.