What is the formation history of the Milky Way? This research aims to shed light on that question through the study of Omega Centauri, a globular cluster composed primarily of old stars. But why this particular cluster? Several features make it one of the most peculiar in our Galaxy.

Omega Centauri (ω Cen) is among the most massive globular clusters in the Milky Way and exhibits a wide spread in stellar metallicity, along with multiple stellar populations that show marked differences in physical and chemical properties. Additionally, it follows a retrograde orbit—rotating in the opposite direction to most of the stars in the Galactic disk—suggesting an extragalactic origin. The cluster also contains stars spanning a broad range of ages and may even host a black hole at its center. These peculiarities set ω Cen apart from the more than 150 globular clusters populating the Galactic halo.

Based on the hypothesis that the Milky Way accreted this former dwarf galaxy, the team used neural networks to compare chemical abundance space between halo stars and those in ω Cen. By analyzing elemental abundances measured in stellar atmospheres—such as carbon, nitrogen, oxygen, magnesium, nickel, among others—the goal was to identify stars with similar chemical signatures, now dispersed across the Galactic halo, that once belonged to the progenitor system of Omega Centauri. This technique, known as chemical tagging, allows astronomers to reconstruct dissolved stellar systems and search for fossil substructures in the halo.

The study relies on data from the Apache Point Observatory Galactic Evolution Experiment (APOGEE), which performs high-resolution spectroscopic mapping of a wide area of the sky to explore the chemical composition of all major stellar populations in the Milky Way.

To train the neural network, the researchers used two data sets: as the positive sample, stars from the core of Omega Centauri; and as the negative sample, populations with distinct chemical evolution histories, such as stars observed by APOGEE in the Magellanic Clouds and the Sagittarius Dwarf Galaxy. “One advantage of neural networks is their ability to analyze multi-dimensional data simultaneously. This allowed us to identify correlations among the various elemental abundances in the sample,” explains Dr. Anguiano of CEFCA.

The study identified 463 high-probability candidate stars in the Galactic halo that are likely remnants of ω Cen, now scattered throughout the Milky Way. “Our findings support the idea that Omega Centauri is a relic of a disrupted dwarf spheroidal galaxy, which has since become part of the Galactic halo,” concludes Dr. Anguiano.

Full article here here.