Australian astronomers have released the chemical fingerprints of almost a million stars, helping piece together the history of the Milky Way galaxy.
The data set is the culmination of 10 years of observations at the Anglo-Australian Telescope (AAT) near Coonabarabran, about 500km west of Sydney in New South Wales. The release of the data coincides with the 50-year anniversary of the facility.
The project is called the GALactic Archaeology with HERMES (GALAH) survey.
Leading GALAH are astronomers from Australia’s ARC Centre of Excellence in All Sky Astrophysics in 3 Dimensions (ASTRO 3D).
“Our work is focused on collecting as much quality data as we can,” says ASTRO 3D’s Sven Buder, a research fellow at the Australian National University.
“GALAH has shown us which chemical elements make up the stars of the Milky Way. This dataset now helps further our ability to accurately age the stars in our neighbourhood and understand where they came from.”
Buder says that the data is a “powerful tool” for scientists to test theories about the formation and history of galaxies and the universe.
The mapping survey includes data from 1.08 million observations of 920,000 galaxies.
“We have measured the elements within these stars, like carbon, nitrogen, oxygen, as well as heavy elements found in our smartphones and electric vehicles,” Buder explains. “This data will help us figure out how these elements are produced in stars, which is fundamental to explaining the origins of the building blocks of life.”
“The GALAH survey has detected signs that some stars may have ‘eaten’ planets that were orbiting them,” adds Daniel Zucker of Macquarie University. “This can be observed by looking at the chemical composition of the star, as the elements from the consumed planet would show up as markers in the star’s spectrum.”
Researchers involved in the project also believe GALAH will help train future artificial intelligence (AI) tools used in astronomy.
“Australia is leading the pack for the future of all of astronomy looking to tackle Big Data,” Zucker says. “This dataset will serve as a one of the leading textbooks for training these AIs.”