spaceSpace and Physics

AI Has Upped Number Of Known Large Protostars From 100 To Over 2,000

Protostars are the precursors to main sequence stars, and can reveal a lot about the formation and early evolution of stars, ESO/L. Calçada

An artificial intelligence (AI) system has proven to be the ultimate “Where’s The Protostar” champion.

These young stars provide a snapshot of the early stages of stellar formation while they are still acquiring mass. Despite their importance in our understanding of the origin of stars in our galaxy, only 100 large protostars (greater than twice the mass of the Sun) had previously been cataloged by scientists. But there are many more out there, rich in information, waiting to be discovered amongst the big data gathered by instruments, such as the Gaia space telescope. This is where AI comes in.


"There is a huge amount of data being produced by Gaia – and AI tools are needed to help scientists make sense of it,” Miguel Vioque, PhD researcher at the University of Leeds, said in a statement. "We are combining new technologies in the way researcher’s survey and map the galaxy with ways of interrogating the mountain of data produced by the telescope – and it is revolutionizing our understanding of the galaxy. This approach is opening an exciting, new chapter in astronomy.”

Vioque and his colleagues used the AI system to probe a small subset (around 4.1 million stars) of Gaia’s data that were likely to contain the desired large protostars, known as Herbig Ae/Be stars. A list of 2,226 stars was generated by the system, with around an 85 percent chance of being the target type. To validate these findings, the team used ground observatories in Spain and Chile to measure the spectra of 145 of the candidate protostars, which confirmed the AI’s precision.

“The results from the ground-based observatories show that the AI tool made very accurate predictions about stars that were likely to fall into the Herbig Ae/Be classification,” Vioque said.

Published in Astronomy and Astrophysics, the authors also described several of the newly discovered protostars that were particularly interesting, including Gaia DR2 428909457258627200. Situated 8,500 light-years away, this 2.3 solar mass protostar has a surface temperature almost 4,000 degrees hotter than the Sun. At only 6 million years old, it is also still in its crucial formation stage that can reveal further secrets of the stars.


Although this study focused on the detection of large protostars, perhaps its greater achievement is the demonstration of AI’s incredible potential in astronomy.

"This research is an excellent example of how the analysis of the Big Data collected by modern scientific instruments, such as the Gaia telescope, will shape the future of astrophysics,” Professor René Oudmaijer, study co-author also from the University of Leeds said. “AI systems are able to identify patterns in vast quantities of data – and it is likely that in those patterns, scientists will find clues that will lead to new discoveries and fresh understanding.”


spaceSpace and Physics