Imagine a massive star locked in a cosmic tango with a black hole, its partner in a deadly dance. One misstep, and boom! The star explodes. That’s precisely what happened in a jaw-dropping discovery made by astronomers using artificial intelligence (AI) to catch supernovae in the act.
Meet SN 2023zkd, a stellar blast unlike anything seen before. First spotted in July 2023, this explosion wasn’t just another twinkle in the sky; it was a cosmic drama unfolding 730 million light-years away. And thanks to a clever AI algorithm, scientists got front-row seats.
The AI, developed by a team led by the University of California, Santa Cruz, scans the skies for unusual flashes in real time.
Ryan Foley, associate professor of astronomy and astrophysics at UC Santa Cruz, said, “Something exactly like this supernova has not been seen before so that it might be very rare. Humans are reasonably good at finding things that ‘aren’t like the others,’ but the algorithm can flag things earlier than a human may notice. This is critical for these time-sensitive observations.”
Once SN 2023zkd lit up the sky, telescopes around the world and in space zoomed in. Two of them, perched atop Haleakalā in Hawaiʻi, are part of the Young Supernova Experiment (YSE), a project that scans 4% of the night sky every three days. That’s like checking 6,000 full moons’ worth of space!
So what caused the blast? The leading theory is that the star was slowly spiraling toward a black hole. As their orbit shrank, the black hole’s gravity stretched the star like cosmic taffy, until it snapped and exploded.
Another possibility? The black hole might’ve gobbled the star entirely, shredding it before it could explode. Either way, the result was the same: one epic light show and a heavier black hole left behind.
Lead author Alexander Gagliano, a fellow at the NSF Institute for Artificial Intelligence and Fundamental Interactions, said, “Our analysis shows that the blast was sparked by a catastrophic encounter with a black hole companion, and is the strongest evidence to date that such close interactions can detonate a star.”
At first, SN 2023zkd looked like a typical supernova, a single bright flash. But then… it got brighter again. That’s weird.
Digging into old data, scientists found the star had been slowly glowing for four years before the explosion. That’s like a candle flickering long before it bursts into flame.
Turns out, the star had been shedding material, gas, and dust into space. When the explosion came, it slammed into this stuff, creating a second, delayed burst of light. Like fireworks hitting fog, the glow lingered and shimmered.
Behind the scenes, this discovery was powered by a mix of AI smarts and human curiosity. Gagliano led the charge, while Foley played the role of skeptical expert, double-checking the data.
At first, the idea that the black hole triggered the supernova almost sounded like science fiction, Foley recalled. So it was essential to make sure all of the observations lined up with this explanation, and Foley said Gagliano methodically demonstrated that they did.
The team’s software platform helped manage observations and integrate AI tools. Co-authors from Harvard, MIT, and UC Santa Cruz brought together expertise in theory, data, and machine learning.
Journal Reference:
- A. Gagliano, V. A. Villar, T. Matsumoto, D. O. Jones. L. Ransome, A. E. Nugent, D. Hiramatsu, K. Auchettl, D. Tsuna, Y. Dong et al. Evidence for an Instability-induced Binary Merger in the Double-peaked, Helium-rich Type IIn Supernova 2023zkd. The Astrophysical Journal. DOI 10.3847/1538-4357/adea38
