Researchers utilized artificial intelligence to construct a three-dimensional representation of an energetic burst, or flare, that erupted near the central black hole of the Milky Way, Sagittarius A* (Sgr A*). This 3D depiction could aid scientists in gaining a clearer understanding of the turbulent surroundings enveloping supermassive black holes in general.
The matter encircling Sgr A* forms a flattened structure known as an "accretion disk," which experiences periodic flares. These flares manifest across various light wavelengths, spanning from high-energy X-rays to low-energy infrared light and radio waves.
Supercomputer simulations propose that a flare observed by the Atacama Large Millimeter/Submillimeter Array (ALMA) on April 11, 2017 originated from two luminous regions of dense material within Sgr A*'s accretion disk, both facing Earth. These luminous regions whirl around the supermassive black hole, which boasts a mass approximately 4.2 million times that of the sun, while separated by roughly half the distance between Earth and the sun, equivalent to around 47 million miles (75 million kilometers).
Crafting a three-dimensional reconstruction of these flares from observational data poses a significant challenge. To address this, the team, led by California Institute of Technology scientist Aviad Levis, proposed a novel imaging technique termed "orbital polarimetric tomography," akin to medical computed tomography (CT) scans commonly conducted in hospitals worldwide.
Levis explained, "The compact vicinity surrounding the galactic center represents an extreme locale where hot, magnetized gas orbits a supermassive black hole at relativistic velocities [speeds nearing that of light]. This distinctive environment fuels highly energetic outbursts known as flares, leaving detectable traces across X-ray, infrared, and radio wavelengths. Recently, theorists have posited several mechanisms for the emergence of such flares, including the sudden formation of exceedingly bright, compact regions within the accretion disk."
He further emphasized that the primary outcome of this endeavor is the recovery of a visual representation depicting the three-dimensional structure of radio brightness surrounding Sgr A* immediately following a flare detection.
Building a black hole from a single pixel
"Sgr A* resides at the core of our own Milky Way galaxy, positioning it as the closest supermassive black hole and an ideal subject for studying such eruptions," remarked Levis. "However, effective study still requires an element of chance, as ALMA observations must align with a flare occurrence."
He recounted that on April 11, 2017, ALMA was surveying Sgr A* immediately following a violent outburst observed in X-rays. The radio data gathered by ALMA exhibited a periodic signal consistent with an orbit around Sgr A*.
"This prompted the development of a computational method capable of extracting the 3D structure from ALMA's time series data," added Levis. "In contrast to the 2D image captured by the Event Horizon Telescope (EHT) of Sgr A*, our focus was on recovering the 3D volume, achieved by applying physical principles governing light propagation along curved paths within a black hole's strong gravitational field."
To attain their findings, researchers delved into physics stemming from Albert Einstein's 1915 theory of gravity, general relativity, and implemented these concepts around supermassive black holes into a neural network. This network then generated the model of Sgr A*.
"This endeavor represents a distinctive collaboration between astronomers and computer scientists, advancing state-of-the-art computational tools from both AI and gravitational physics," stated Levis. "Each field contributed significantly to this initial effort to unveil the 3D radio emission structure around Sgr A*."
He clarified that the outcome wasn't a conventional photograph but rather a computational 3D image derived from time-series observations, where a neural network was constrained by the anticipated physics of gas orbiting the black hole and emitting synchrotron radiation.
Levis elucidated that the team computationally positioned 3D "emissions" orbiting Sgr A* and refined the structure until it matched observations using neural radiance fields and general relativistic ray tracing.
"This methodology allowed us to draw conclusions about Sgr A*'s environment in accordance with theory, revealing brightness concentrated in small regions within the accretion disk," noted Levis. Nonetheless, certain aspects of their work surprised the team.
"The most notable surprise was our ability to reconstruct the 3D structure from light curve observations—a video essentially captured by a single flickering pixel," Levis expressed. "Consider this: recovering a video from just one pixel seems practically impossible. However, our success lies in reconstructing the emission's 3D structure around a black hole, leveraging gravitational and emission physics."
He emphasized that ALMA's measurement of both light intensity and polarization provided invaluable insights into the 3D structure of flares around Sgr A*.
Looking ahead, Levis disclosed plans to adjust the simulation's physics parameters constrained by AI.
"These findings represent an exciting initial stride, contingent upon the assumption that Sgr A* behaves as expected based on gravitational and emission models," concluded Levis. "Future endeavors aim to relax these constraints, permitting deviations from anticipated physics. Our interdisciplinary approach paves the way for probing new questions, furthering our comprehension of black holes and the cosmos."
