The unusual system of three "super puff" planets has at least one more planet, revealed by its gravitational tug on other planets
An unusual planetary system with three known ultra-low density "super-puff" planets has at least one more planet, according to new research led by researchers from Penn State and Osaka University. The research team set out to study Kepler-51d, the third planet in the system, with NASA's James Webb Space Telescope (JWST) but almost missed their chance when the planet unexpectedly passed in front of its star two hours earlier than models predicted. After scrutinizing new and archival data from a variety of space and Earth-based telescopes, the researchers found that the best explanation is the presence of a fourth planet, whose gravitational pull impacts the orbits of the other planets in the system.
The new planet's discovery is detailed in a paper appearing Dec. 3 in the Astronomical Journal.
"Super puff planets are very unusual in that they have very low mass and low density," said Jessica Libby-Roberts, Center for Exoplanets and Habitable Worlds Postdoctoral Fellow at Penn State and co-first author of the paper. "The three previously known planets that orbit the star, Kepler-51, are about the size of Saturn but only a few times the mass of Earth, resulting in a density like cotton candy. We think they have tiny cores and huge atmospheres of hydrogen of helium, but how these strange planets formed and how their atmospheres haven't been blown away by the intense radiation of their young star has remained a mystery. We planned to use JWST to study one of these planets to help answer these questions, but now we have to explain a fourth low-mass planet in the system!"
When a planet passes in front of -- or transits -- its star when viewed from Earth, it blocks some of the star's light, causing a slight decrease in the star's brightness. The duration and amount of that decrease gives clues to the planet's size and other characteristics. Planets transit when they complete an orbit around their star, but sometimes they transit a few minutes early or late because the gravity from other planets in the system tug on them. These minor differences are known as transit timing variations and are built into astronomers' models to allow them to accurately predict when planets will transit.
The researchers said they had no reason to believe the three-planet model of the Kepler-51 system was inaccurate, and they successfully used the model to predict the transit time of Kepler-51b in May 2023 and followed-up with the Apache Point Observatory (APO) telescope to observe it on schedule.
"We also tried to use the Penn State Davey Lab telescope to observe a transit of Kepler-51d in 2022, but some poorly timed clouds blocked our view right as the transit was predicted to start," Libby-Roberts said. "It's possible we could have learned something was off then, but we had no reason to suspect that Kepler-51d wouldn't transit as expected when we planned to observe it with JWST."
The team's three-planet model predicted that Kepler-51d would transit around 2 a.m. EDT in June 2023, and the researchers prepared to observe the event with both JWST and APO.
"Thank goodness we started observing a few hours early to set a baseline, because 2 a.m. came, then 3, and we still hadn't observed a change in the star's brightness with APO," Libby-Roberts said. "After frantically re-running our models and scrutinizing the data we discovered a slight dip in stellar brightness immediately when we started observing with APO, which ended up being the start of the transit -- 2 hours early, which is well beyond the 15-minute window of uncertainty from our models!"
When the researchers analyzed the new APO and JWST data, they confirmed that they had captured the transit of Kepler-51d, albeit considerably earlier than expected.
"We were really puzzled by the early appearance of Kepler-51d, and no amount of fine-tuning the three-planet model could account for such a large discrepancy," said Kento Masuda, associate professor of earth and space science at Osaka University and co-first author of the paper. "Only adding a fourth planet explained this difference. This marks the first planet discovered by transit timing variations using JWST."
To help explain what is happening in the Kepler-51 system, the research team revisited previous transit data from NASA's Kepler space telescope and NASA's Transiting Exoplanet Survey Satellite (TESS). They also made new observations of the inner planets in the system, including with the Hubble Space Telescope and the California Institute of Technology's Palomar Observatory telescope, and obtained archival data from several ground-based telescopes. Because the new planet, Kepler-51e, has not yet been observed transiting -- perhaps because it may not pass in the line of sight between its star and Earth -- the researchers noted how important it was to obtain as much data as possible to support their new models.
"We conducted what is called a 'brute force' search, testing out many different combinations of planet properties to find the four-planet model that explains all of the transit data gathered over the past 14 years," Masuda said. "We found that the signal is best explained if Kepler-51e has a mass similar to the other three planets and follows a fairly circular orbit of about 264 days -- something we would expect based on other planetary systems. Other possible solutions we found involve a more massive planet on a wider orbit, though we think these are less likely."
Accounting for a fourth planet and adjusting the models also changes the expected masses of the other planets in the system. According to the researchers, this impacts other inferred properties about these planets and informs how they might have formed. Although the inner three planets are slightly more massive than previously thought, they are still classified as super puffs. However, it is unclear if Kepler-51e is also a super puff planet, because the researchers have not observed a transit of Kepler-51e and therefore cannot calculate its radius or density.
"Super puff planets are fairly rare, and when they do occur, they tend to be the only one in a planetary system," Libby-Roberts said. "If trying to explain how three super puffs formed in one system wasn't challenging enough, now we have to explain a fourth planet, whether it's a super puff or not. And we can't rule out additional planets in the system either."
Because the researchers believe Kepler-51e has an orbit of 264 days, they said that additional observing time is needed to get a better picture of the impacts of its gravity -- or that of additional planets -- on the inner three planets in the system.
"Kepler-51e has an orbit slightly larger than Venus and is just inside the star's habitable zone, so a lot more could be going on beyond that distance if we take the time to look," Libby-Roberts said. "Continuing to look at transit timing variations might help us discover planets that are further away from their stars and might aid in our search for planets that could potentially support life."
The researchers are currently analyzing the rest of the JWST data, which could provide information about the atmosphere of Kepler-51d. Studying the composition and other prosperities the inner three planets could also improve understanding how unusual ultra-low density super puff planets formed, the researchers said.
In addition to Libby-Roberts and Masuda, who led the Kepler-51d team, the international research team includes John Livingston at the National Astronomical Observatory of Japan, who coordinated most of the ground-based follow-ups; many ground-based observers; the Kepler-51b team; and the Palomar team. See the article for a full list of authors and affiliations.
NASA supported this research through JWST and Hubble Space Telescope grants. Computations for this research were performed on the Penn State's Institute for Computational and Data Sciences Advanced CyberInfrastructure.