(ORDO NEWS) — Astronomers at the California Institute of Technology have used a machine learning algorithm to classify 1,000 supernovae completely offline.
The algorithm was applied to data acquired by the Zwicky Transient Facility (ZTF), a sky survey instrument based at Caltech‘s Palomar Observatory.
“We needed help, and we knew that once we got our computers to do the job, they would take a big load off us,” says Kristoffer Fremling, Caltech staff astronomer and one of the creators of a new algorithm called SNIascore.
“SNIascore classified its first supernova in April 2021 and a year and a half later we hit a good milestone of 1,000 supernovae.”
ZTF scans the night sky every night for changes called transient events. This concept includes everything from moving asteroids to exploding stars known as supernovae.
ZTF sends out hundreds of thousands of alerts per night to astronomers around the world, notifying them of these events. Astronomers then use other telescopes to track and explore the nature of changing objects.
But the members of the ZTF team can’t make sense of the sheer amount of data coming in every night on their own.
A team of scientists have developed machine learning algorithms to help with data processing. They created the SNIAscore, designed to classify supernova candidates.
Supernovae fall into two broad classes: type I and type II. Type I supernovae are devoid of hydrogen, while Type II supernovae are rich in hydrogen.
The most common type I supernova occurs when a massive star steals matter from a companion star, eventually causing a thermonuclear explosion. A Type II supernova occurs when a massive star collapses under its own gravity.
SNIascore can currently classify Type Ia supernovae, and Fremling and colleagues are actively working to extend the algorithm’s ability to classify other types of supernovae in the near future.
Contact us: [email protected]