Astronomers have discovered hundreds of "polluted" white dwarf stars in our Milky Way galaxy, actively consuming planets in their orbits. These stars provide invaluable insights into the composition of these distant, destroyed planets. However, finding them is a challenging task.
Traditionally, astronomers have relied on manually examining vast amounts of survey data for signs of these stars, followed by further observations to confirm their suspicions. This method is time-consuming and inefficient.
A team led by University of Texas at Austin graduate student Malia Kao has developed a revolutionary approach using a form of artificial intelligence called manifold learning. This AI technique has significantly accelerated the process, achieving a remarkable 99% success rate in identifying these "polluted" stars. The results of their research were published in The Astrophysical Journal on July 31.
White dwarfs represent the final stage of a star's life. Having exhausted their fuel, they shed their outer layers and slowly cool. Our own sun is destined to become a white dwarf in about 6 billion years.
Occasionally, planets orbiting a white dwarf are drawn in by its powerful gravity, torn apart, and consumed. This process "pollutes" the star with heavy metals from the planet's interior. As white dwarf atmospheres are predominantly composed of hydrogen and helium, the presence of other elements is a clear indicator of external sources.
"For polluted white dwarfs, the inside of the planet is literally being seared onto the surface of the star for us to look at," Kao explained. "They are the best way we can currently characterise planetary interiors."
Co-author Keith Hawkins, an astronomer at UT, added, "Finding these polluted white dwarfs is crucial because they offer the only authentic way to understand the composition of planets beyond our solar system."
The subtle traces of polluting metals in these stars' atmospheres make them difficult to detect. Moreover, astronomers have a limited window of opportunity to find them.
The team utilised data from the Gaia space telescope, one of the largest spectroscopic surveys of white dwarfs to date. Despite the low resolution of the data, they successfully applied AI to identify these elusive stars.
Their AI approach, manifold learning, groups similar features in a dataset, creating a visual chart that helps researchers identify promising candidates for further investigation.
The astronomers developed an algorithm to sift through over 100,000 potential white dwarfs. A cluster of 375 stars stood out, exhibiting the characteristic presence of heavy metals in their atmospheres. Subsequent observations using the Hobby-Eberly Telescope at UT's McDonald Observatory confirmed their findings.
Kao highlighted the significance of their discovery: "Our method can increase the number of known polluted white dwarfs tenfold, enabling us to better study the diversity and geology of planets outside our solar system. Ultimately, we strive to determine whether life can exist beyond our solar system. If ours is unique among planetary systems, it might also be unique in its ability to sustain life."
This innovative approach demonstrates the transformative potential of artificial intelligence in scientific research. To celebrate and showcase these advancements, UT Austin has declared 2024 the Year of AI.
This research utilised data from the European Space Agency (ESA) mission Gaia, processed by the Gaia Data Processing and Analysis Consortium. Follow-up observations were conducted using the Hobby-Eberly Telescope (HET), a collaborative project involving UT Austin, Pennsylvania State University, Ludwig Maximilians-Universitaet Muenchen, and Georg-August Universitaet Goettingen, and the Very Large Telescope (VLT) at the European Southern Observatory (ESO). The Texas Advanced Computing Center at UT Austin provided high-performance computing, visualisation, and storage resources for this research.
This groundbreaking research has opened a new window into the composition and evolution of planets beyond our solar system. It exemplifies the power of artificial intelligence to solve scientific mysteries and contribute to our understanding of the universe.