Artificial Intelligence Helps Find Stars That Devour Planets

Aug. 2 () –

Through multiple learninga new form of artificial intelligence, astronomers have achieved a 99 percent success rate in Identification of ‘elusive’ stars that devour planets.

Hundreds of “polluted” white dwarf stars have recently been discovered in our Milky Way galaxy. These are white dwarfs that are actively consuming planets in their orbit. They are a valuable resource for studying the interiors of these distant, demolished planets. They are also hard to find.

Historically, astronomers have had to manually sift through mountains of survey data looking for signals from these stars. Follow-up observations would prove or disprove their suspicions. The new technique, applied by a team led by University of Texas (UT) at Austin graduate student Malia Kao, has dramatically sped up the process. The findings are published in The Astrophysical Journal.

White dwarfs are stars in their final stages of life. They have exhausted their fuel, released their outer layers into space, and are slowly cooling down. One day, our sun will become a white dwarf, but that won’t be for another 6 billion years.

Planets orbiting a white dwarf are sometimes pulled in by their star’s gravity, torn apart, and consumed. When this happens, the star becomes “polluted” with heavy metals from the planet’s interior. Because the atmospheres of white dwarfs are composed almost entirely of hydrogen and helium, the presence of other elements can be reliably attributed to external sources.

“In the case of polluted white dwarfs, the interior of the planet is literally being burned onto the surface of the star so that we can observe it,” Kao said. it’s a statement“Polluted white dwarfs are currently the best way to characterize planetary interiors.”

“Put another way,” added Keith Hawkins, a UT astronomer and co-author of the paper, “it’s the only real way to figure out what planets outside the solar system are made of, “which means that finding these contaminated white dwarfs is critical.”

Unfortunately, evidence of these stars, which are identified by the contaminating metals in their atmospheres, can be subtle and difficult to detect. And astronomers must find them within a relatively short time frame.

Although astronomers can identify these stars by manually reviewing data from astronomical surveys, this can be time-consuming. To test a faster process, the team applied AI to data available from the Gaia space telescope. “Gaia provides one of the largest spectroscopic surveys of white dwarfs to date, But the data are so low resolution that we think it would not be possible to find white dwarfs contaminated with it.“Hawkins said. “This work shows that it can be done.”

To find these elusive stars, the team used an AI technique called manifold learning. With it, an algorithm looks for similar features in a data set and groups similar elements into a simplified visual graph. Then, Researchers can review the graph and decide which clusters warrant further investigation.

AN ALGORITHM CLASSIFIED MORE THAN 100,000 POSSIBLE WHITE DWARFS

Astronomers created an algorithm to classify more than 100,000 possible white dwarfs. Of these, A group of 375 stars looked promising: They displayed the key characteristic of having heavy metals in their atmospheres. Follow-up observations with the Hobby-Eberly Telescope at UT’s McDonald Observatory confirmed the astronomers’ suspicions.

“Our method can increase the number of known polluted white dwarfs tenfold, which will allow us to better study the diversity and geology of planets outside our solar system,” Kao said. “Ultimately, we want to determine whether life can exist outside our solar system. If ours is unique among planetary systems, It may also be unique in its ability to sustain life.”

This investigation used data from the European Space Agency’s Gaia mission (ESA). The data were processed by the Gaia Data Processing and Analysis Consortium.

Follow-up observations were obtained with the Hobby-Eberly Telescope (HET), which is a joint project of the University of Texas at Austin, Pennsylvania State University, Ludwig Maximilians-Universitaet Muenchen and Georg-August-University Goettingen, and with the Very Large Telescope (VLT) of the European Southern Observatory (ESO).

The Texas Advanced Computing Center at UT Austin provided high-performance computing, visualization, and storage resources for this research.