Machine learning has sharpened up the first direct photo ever taken of a black hole, revealing a wealth of exciting new details.
In 2019, the Event Horizon Telescope (EHT) collaboration released an image of the supermassive black hole at the heart of the galaxy M87, which lies about 55 million light-years from Earth. The photo shows the black hole, known as M87*, surrounded by a fuzzy ring of light, which is emitted by fast-moving gas and dust falling into its maw.
That ring isn’t so fuzzy anymore, thanks to a new machine-learning tool known as PRIMO, a new study reports.
“Since we cannot study black holes up close, the detail in an image plays a critical role in our ability to understand its behavior,” study lead author Lia Medeiros, of the Institute for Advanced Study in New Jersery, said in a statement. “The width of the ring in the image is now smaller by about a factor of two, which will be a powerful constraint for our theoretical models and tests of gravity.”
Related: Black holes of the universe (images)
The EHT team created the 2019 M87* image using a network of seven radio telescopes at different spots around the globe. They linked up the data gathered by these scopes using a technique known as interferometry, essentially creating a virtual instrument the size of Earth.
There were still gaps in the data, however — and that’s where PRIMO (short for “principal-component interferometric modeling”) comes in. The study team trained the algorithm by exposing computers to more than 30,000 simulated images of feeding black holes. PRIMO found common patterns, then applied this newfound knowledge to generate the new image using the EHT’s M87* dataset.
“PRIMO is a new approach to the difficult task of constructing images from EHT observations,” study co-author Tod Lauer, a scientist at the U.S. National Science Foundation’s NOIRLab, said in the same statement.Â
“It 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,” Lauer added.
PRIMO’s new image should help astronomers nail down the mass of M87*, which is thought to be about 6.5 billion times that of Earth’s sun, as well as some other characteristics, study team members said.
The new machine-learning technique could also be applied to other observations, such as EHT’s 2022 image of Sagittarius A*, the supermassive black hole at the heart of our own Milky Way galaxy.
“The 2019 image was just the beginning,” Medeiros said. “If a picture is worth a thousand words, the data underlying that image have many more stories to tell. PRIMO will continue to be a critical tool in extracting such insights.”
The new study (opens in new tab) was published online today (April 13) in The Astrophysical Journal Letters.
Mike Wall is the author of “Out There (opens in new tab)” (Grand Central Publishing, 2018; illustrated by Karl Tate), a book about the search for alien life. Follow him on Twitter @michaeldwall (opens in new tab). Follow us on Twitter @Spacedotcom (opens in new tab) or on Facebook (opens in new tab). Â
