We now have a better image of black hole M87, thanks to machine learning

This has important implications for measuring the mass of the central black hole in M87.
Sejal Sharma
Old (left) vs new (right) images of the black hole M78
Old (left) vs new (right) images of the black hole M78

NoirLab 

Look at the image on the left and then the image on the right. They are by no means identical. But what if we told you that both the images are of the same object?

The object being a supermassive black hole.

The images were taken four years apart. Scientists have used a machine learning technique to sharpen the image of the black hole at the center of a galaxy called Messier 87 or M87. Previously captured in the year 2019 by the Event Horizon Telescope (EHT) collaboration, the image was so blurry that an astrophysicist who was part of the research team called it a "fuzzy orange donut." 

A surprisingly narrow outer ring

The latest image has been generated by a new machine-learning technique known as PRIMO, which used the same data set of 2019. The new portrait reveals a clearer thin halo of yellow gas, giving it a bit more structure visually. But it is still a blurry image.

PRIMO (principal-component interferometric modeling) was developed by members of the EHT, who also published their research paper in The Astrophysical Journal Letters. PRIMO allowed the researchers to recover high-fidelity images of the black hole in the presence of sparse coverage.

The new image not only illustrates the full extent of the black hole’s central dark region but will also help in the accurate measurement of its width and diameter.

The width, as can be seen in the new image, is almost half of how it looked in the old image. 

In a statement published Thursday, it was revealed that the EHT collaboration back in 2017 used a wide network of seven different radio telescopes at various locations in the world. This was done to create an Earth-sized virtual telescope which could help them in observing the spherical outer boundary of the black hole.

Though the virtual telescope was remarkable in that it allowed the researchers to see the finer details of the black hole, it lacked the capability and the power of an actual earth-sized telescope. PRIMO helped in filling those gaps.

“With our new machine-learning technique, PRIMO, we were able to achieve the maximum resolution of the current array,” says lead author Lia Medeiros. “Since we cannot study black holes up close, the detail in an image plays a critical role in our ability to understand its behavior. 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.” 

The researchers believe that further EHT observations will allow improvements in the effective resolution of the image and to constrain the blurriness.

Add Interesting Engineering to your Google News feed.
Add Interesting Engineering to your Google News feed.
message circleSHOW COMMENT (1)chevron
Job Board