The Milky Way is our galactic home, containing our solar system and continuing into a giant band of densely packed stars that stretches across clear night skies around the world – but, it turns out, not all of those stars always called our galaxy home. New research by Lina Necib, a postdoctoral theoretical physics researcher at Caltech, applied supercomputing power to discover a cluster of stars in the Milky Way that originated outside the galaxy.
The road to the discovery began with two key elements: a project called Feedback In Realistic Environments (or FIRE), a cross-university collaboration working to develop detailed simulations of galaxies; and the Gaia space observatory, which the European Space Agency opened in 2013 with the aim of precisely mapping a billion stars, including those in our galaxy. The FIRE simulations were completed across months on a series of supercomputers, including Blue Waters at the National Center for Supercomputing Applications (NCSA) and Stampede2 at the Texas Advanced Computing Center (TACC).
The researchers, led by Necib, combined data from these projects to study the kinematics (motions) of stars as they formed the modern Milky Way. To do this, they used the Gaia mock catalogues, which assessed how the simulated galaxies produced by FIRE would be observed by Gaia if they were real. By analyzing the star motions in the mock catalogues with deep learning on high-performance clusters at the University of Oregon, they developed a method for categorizing stars as either original to the galaxy or obtained through galactic mergers.
Then, they tackled the real world (or rather, the real galaxy) using data from Gaia. “We asked the neural network, ‘Based on what you’ve learned, can you label if the stars were accreted or not?'” Necib said in an interview with Aaron Dubrow at TACC. The network then made its best guesses, assigning values between 0 (zero confidence that a star was born outside the Milky Way) to 1 (full confidence). The model was able to correctly identify a merged star formation known as the “Gaia sausage,” validating the accuracy of the approach.
Then, they found something.
The neural network identified a cluster of 250 stars in the Milky Way’s disk with motion that indicated that they were the product of a merger. “Your first instinct is that you have a bug,” Necib said. “And you’re like, ‘Oh no!’ So, I didn’t tell any of my collaborators for three weeks. Then I started realizing it’s not a bug, it’s actually real and it’s new.”
Necib named the cluster Nyx, after the Greek goddess of the night.
“Galaxies form by swallowing other galaxies,” Necib said. “We’ve assumed that the Milky Way had a quiet merger history, and for a while it was concerning how quiet it was because our simulations show a lot of mergers. Now, with access to a lot of smaller structures, we understand it wasn’t as quiet as it seemed. It’s very powerful to have all these tools, data and simulations. All of them have to be used at once to disentangle this problem. We’re at the beginning stages of being able to really understand the formation of the Milky Way.”
In the wake of their success, Necib and her colleagues are continuing their research into galactic formation on the powerful Frontera system at TACC. They’re also using ground-based telescopes to take a deeper dive into the Nyx cluster in the hopes of understanding when and how Nyx entered into the Milky Way.
Header image: a still from a galaxy formation simulation. Image courtesy of Caltech.
To read the TACC article discussing this research, click here.