A team of researchers from Google Brain, the University of Texas at Austin, MIT, the University of California, Berkeley, Harvard-Smithsonian Center for Astrophysics, NASA, and NOAO has used artificial intelligence (AI) to find two more exoplanets in data from NASA’s Kepler Space Telescope’s extended mission, K2.
An artist’s impression of the super-Earth K2-265b. Image credit: Sci-News.com.
The newly-discovered planets, designated K2-293b and K2-294b, are both super-Earths, with sizes of 2.45 and 1.66 times that of our planet.
K2-293b orbits K2-293 (also known as EPIC 246151543), a star about 1,300 light-years away in the constellation of Aquarius.
The other planet is in orbit around K2-294 (EPIC 246078672), a star approximately 1,230 light-years away, also located in Aquarius.
K2-293b is probably a ‘puffy’ planet with a volatile atmosphere. It has an orbital period of 13.1 days, so it is strongly irradiated by its host star.
K2-294b is likely still rocky because it is probably too close to its host star to have a hydrogen/helium atmosphere. Given its very short orbital period of 2.5 days, this planet is not Earth-like; instead it is heated by its host star to scorching temperatures.
“The two planets we found are really close in to their host star, they have short orbital periods, and they’re hot. They are slightly larger than Earth,” said Anne Dattilo, an undergraduate student at the University of Texas at Austin.
To find the planets, Dattilo and co-authors created an algorithm that sifts through the K2 data to ferret out signals that were missed by traditional planet-hunting methods.
“K2 data is more challenging to work with because the spacecraft is moving around all the time,” said Dr. Andrew Vanderburg, also from the University of Texas at Austin.
“This change came about after a mechanical failure. While mission planners found a workaround, the spacecraft was left with a wobble that AI had to take into account.”
Once the astronomers used their algorithm to find K2-293b and K2-294b, they followed up by studying the host stars using ground-based telescopes to confirm that the planets are real.
These observations were done with the 1.5-m telescope at the Smithsonian Institution’s Whipple Observatory in Arizona and the Gillett Telescope at Gemini Observatory in Hawaii.
“AI will help us search the data set uniformly,” Dr. Vanderburg said.
“Even if every star had an Earth-sized planet around it, when we look with Kepler, we won’t find all of them. That’s just because some of the data’s too noisy, or sometimes the planets are just not aligned right.”
“So, we have to correct for the ones we missed. We know there are a lot of planets out there that we don’t see for those reasons.”
The team’s work will be published in the Astronomical Journal.
_____
Anne Dattilo et al. 2019. Identifying Exoplanets with Deep Learning II: Two New Super-Earths Uncovered by a Neural Network in K2 Data. AJ, in press; arXiv: 1903.10507
