A team of astronomers at the California Institute of Technology has made a groundbreaking discovery using neural network and three-dimensional (3D) technology, akin to CT scans, to perform “computed tomography” (CT) on black holes. This innovative approach has allowed them, for the first time, to reconstruct a map of high-energy burst events near Sagittarius A*.
The research results were published in “Nature Astronomy” on the 22nd, providing a clearer picture of how flares around black holes are formed.
Sagittarius A*, an extremely bright and complex radio wave source located in the center of the Milky Way, is the closest supermassive black hole to Earth and is considered the best target for studying black hole physics.
Previously, the Event Horizon Telescope (EHT), a global network of eight radio telescopes, had managed to capture the first-ever image of a black hole, located in the galaxy M87 with Sagittarius A* as the subject. This was a monumental achievement that provided visual confirmation of these enigmatic entities. Since then, scientists have been tracking high-energy outbursts from Sagittarius A*.
A New Imaging Technology
Traditionally, studying black holes has relied on observing their influence on surrounding matter. The immense gravity of a black hole warps spacetime, causing stars and gas to orbit at breakneck speeds or be devoured entirely. By analyzing these motions, astronomers can estimate the mass and spin of the black hole.
However, this indirect approach leaves much about a black hole’s internal makeup shrouded in mystery.
To tackle this challenge the research team proposed a new imaging technology, similar to that used in medical CT, which they named “orbital polarization tomography.”
The process of tomography involves imaging by sections or sectioning, through the use of any kind of penetrating wave. In the context of black holes, this means observing and analyzing the event horizon — the boundary marking the limits of a black hole.
This technology was then used to study the 3D appearance of flares at radio wavelengths, using observations from the Atacama Large Millimeter/submillimeter Array (ALMA) on April 11, 2017.
Reconstructing 3D images from this data set presented challenges due to distance and brightness changes in particle details. To overcome these difficulties, the research team used a new computer technology based on neural networks and used the predicted physical properties of black holes and electromagnetic radiation processes to constrain the neural network.
The resulting image shows that the flare may have originated from two bright spots on the accretion disk, which is almost directly facing the Earth. These bright spots rotate clockwise around the black hole, with an orbital radius half the distance between the Earth and the Sun (about 75 million kilometers).
The reconstructed flare structure is similar to previous computer simulations, verifying people’s general understanding of the extreme environment around the black hole.
