Scientists may have discovered an unprecedented type of supernova after scanning the sky for strange activities using artificial intelligence (AI) like Spotify.
The AI unearthed signs of what could have been a giant star, just as it was trying to swallow a nearby black hole.
The stellar explosion, known as the SN 2023ZKD, was discovered in July 2023 at the Zwicky Transient facility, a full-skied astronomical survey based at the Palomar Observatory in California. However, Zwicky did not find a random explosion. Rather, they were led to the right place using optimized algorithms to find strange night-like activities.
You might like it
Discovering supernova signs early is the key to catching how supernovas begin, evolve and disappear, and provides insight into how these explosions work.
In this case, the AI discovered an unusually bright discovery a few months before the explosion occurred. Alex Galliano, a postdoctoral researcher at the Institute of AI and Fundamental Interactions, and Ashley Villard, a Supernova researcher and assistant professor for Astrophys at the Harvard Smithsonian Center, spoke to Live Science in Science.
This quick alert allowed many large observatory to take part in the action and provide observations of large wavelengths.
Related: 2 “New Stars” exploded into the night sky at once – potentially for the first time in history
There are several ideas about what these telescopes actually saw, but the scientists behind the new research say the explosion is most likely from a giant star orbiting a black hole. The separation between these two objects was reduced as they were pulled against each other. Eventually, the star attempted to consume a black hole due to gravity stress, and exploded in the process.
Alternatively, the black hole may have shredded the stars through a process known as “spaghetification,” causing the explosion, but the data does not strongly suggest that, Galliano said.
By looking at the chemical composition of the giant stars, the team also discovered that they hadn’t lost all of the outermost material before it exploded.
“This suggests that binary interactions are far more troubling than astronomers thought,” Gagliano said. “ Future events will show you how giant star explosions are shaped by companion interaction.
Gagliano warned that he had never seen enough of these explosions to fully predict how giant stars and black holes interact. However, data is “very difficult to explain without a binary system,” meaning that black holes and stars are very likely to be involved in some way.
AI support
The AI used in discovery is called LightCurve’s anomaly identification and similarity search (LAISS). As Astronomical AI is based on the Spotify algorithm, LAISS recommends astronomical observations in a similar way that will lead you to songs that Spotify users may enjoy.
The latest explosion attracted the attention of LAISS due to its properties from the light of the binary system and its 730 million light-year location from Earth. The features of the SN 2023ZKD were “compared with a large reference dataset of known objects to identify statistical outliers,” Gagliano said. “Anomalous signals can indicate rare or previously invisible phenomena.”
When Laiss finds something interesting, they flag Slack bots, instant messaging services and candidates and post them on dedicated channels, allowing team members to check their findings in real time.
“This streamlined system allows astronomers to quickly target the most promising and rare findings,” Galliano said.
After the explosion, the light pattern on the SN 2023ZKD became very strange. At first it brightened like a typical supernova and then declined. But astronomers really began to pay attention when they brightened up again. Archive data showed more strange behavior. The star, which had been consistently bright for a while, gradually brightened up after four years of explosion.
Astronomers believe that light comes from the excess material that the stars had fallen off. Initially, shock waves from the supernova tilled by the lower density of gases in the area brightened. Another peak of brightness came as the shock waves followed the dust clouds.
As for the existence of black holes, astronomers have inferred it from both the brighter and gas and dust structures of the strange stars from the years before the explosion.
Laiss helped astronomers to see all this extra detail. “If humans had waited until they flagged the 2023ZKD, they would have missed the early signatures of the surrounding disks and the presence of black hole companions. AI systems like LAISS will help them regularly find rare explosions without relying on luck, and at a stage sufficient to discover their origins.
The results were published in the Astrophysical Journal on Wednesday (August 13th).
Black Hole Quiz: How Massive is your knowledge of the universe?
Source link
