Two new exoplanets have been discovered thanks to NASA’s collaboration with Google’s artificial intelligence (AI). One of those in today’s announcement is an eighth planet – Kepler-90i – found orbiting the Sun-like star Kepler-90. This makes it the first system discovered with an equal number of planets to our own Solar system.
A mere road trip away, at 2,545 light-years from Earth, Kepler-90i orbits its host star every 14.4 Earth days, with a sizzling surface temperature similar to Venus of 426°C.
The new exoplanets are added to the growing list of known worlds found orbiting other stars.
This new Solar system rival provides evidence that a similar process occurred within Kepler-90 that formed our very own planetary neighbourhood: small terrestrial worlds close to the host star, and larger gassy planets further away. But to say the system is a twin of our own Solar system is a stretch.
The entire Kepler-90 system of eight planets would easily fit within Earth’s orbit of the Sun. All eight planets, bar Kepler-90h, would be too hostile for life, lying outside the so-called habitable zone.
Evidence also suggests that planets within the Kepler-90 system started out farther apart, much like our own Solar system. Some form of migration occurred, dragging this system inwards, producing the orbits we see in Kepler-90 today.
Google’s collaboration with NASA’s space telescope Kepler mission has now opened up new and exciting opportunities into AI helping with scientific discoveries.
So how exactly did Google’s AI discover these planets? And what sort of future discoveries can this technology provide?
Traditionally, software developers program computers to perform a particular task, from playing your favourite cat video, to determining exoplanetary signals from space based telescopes such as NASA’s Kepler Mission.
These programs are executed to serve a single purpose. Using code intended for cat videos to hunt exoplanets in light curves would lead to some very interesting, yet false, results.
Googles’s AI is programmed rather differently, using machine learning. In machine learning, AI is trained through artificial neural networks – somewhat replicating our brain’s biological neural networks – to perform tasks like reading this article. It then learns from its mistakes, becoming more efficient at its particular task.
Google’s DeepMind AI, AlphaGo, was trained previously to play Go, an extremely complex yet elegant Chinese board game. Last year, AlphaGo defeated Lee Sedol, the world’s best Go player, by four games to one. It simply trained itself by watching thousands of previously played games, then competing against itself.
In our exoplantary case, AI was trained to identify transiting exoplanets, sifting through 15,000 signals from the Kepler exoplanet catalogue. It learned what was and wasn’t a signal caused by an exoplanet eclipsing its host star. These 15,000 signals were previously vetted by NASA scientists prior to the AI’s training, guiding it to a 96% efficiency of detecting known exoplanets.
Researchers then directed their AI network to search in multiplanetary systems for weaker signals. This research culminated in today’s announcement of both Kepler-90i and another Earth-sized exoplanet, Kepler-80g, in a separate planetary system.
Google’s AI has analysed only 10% of the 150,000 stars NASA’s Kepler Mission has been eyeing off across the Milky Way galaxy.
There’s potential then for sifting through Kepler’s entire catalogue and finding other exoplanetary worlds that have either been skimmed by scientist or haven’t been checked yet, due to Kepler’s rich data set. And that’s exactly what Google’s researchers are planning to do.
Machine learning neural networks have been assisting astronomers for a few years now. But the potential for AI to assist in exoplanetary discoveries will only increase within the next decade.
The Kepler mission has been running since 2009, with observations slowly coming to an end. Within the next 12 months, all of its on-board fuel will be fully depleted, ending what has been, one of the greatest scientific endeavours in modern times.
Kepler’s successor, the Transiting Exoplanet Survey Satellite (TESS) will be launching this coming March.
TESS is predicted to find 20,000 exoplanet candidates during its two-year mission. To put that into perspective, in the past 25 years, we’ve managed to discover just over 3,500.
This unprecedented inundation of exoplanetary data needs to either be confirmed by other transiting observations or other methods such as ground-based radial velocity measurements.
There just isn’t enough people-power to sift through all of this data. That’s why these machine learning networks are needed, so they can aid astronomers in sifting through big data sets, ultimately assisting in more exoplanetary discoveries. Which begs the question, who exactly gets credit for such a discovery?
Jonti Horner, University of Southern Queensland; Jake Clark, University of Southern Queensland; Rob Wittenmyer, University of Southern Queensland, and Stephen Kane, University of California, Riverside
The discovery of a planet with a highly elliptical orbit around an ancient star could help us understand more about how planetary systems form and evolve over time.
The new planet, HD76920b, is four times the mass of Jupiter, and can be found some 587 light years away in the southern constellation Volans, the Flying Fish. At its closest it skims the surface of its host star, HD76920. At its furthest, it orbits almost twice as far from its star as Earth does from the Sun.
Details of the planet and its discovery are published today. So how does this fit into the planet formation narrative, and are planets like it common in the cosmos?
Before the first exoplanet discovery, our understanding of how planetary systems formed came from the only example we had at the time: our Solar system.
Close to the Sun orbit four rocky planets – Mercury, Venus, Earth and Mars. Further out are four giants – Jupiter, Saturn, Uranus and Neptune.
The eight planets move in almost circular orbits, close to the same plane. The bulk of the debris also lies close to that plane, although on orbits that are somewhat more eccentric and inclined.
How did this system form? The idea was that it coalesced from a disk of material surrounding the embyronic Sun. The colder outer reaches were rich in ices, while the hotter inner regions contained just dust and gas.
Over millions of years, the tiny particles of dust and ice collided with one another, slowly building ever larger objects. In the icy depths of space, the giant planets grew rapidly. In the hot, rocky interior, growth was slower.
Eventually, the Sun blew away the gas and dust leaving a (relatively) orderly system – roughly co-planar planets, moving on near-circular orbits.
The first exoplanets, discovered in the 1990s, shattered this simple model of planet formation. We quickly learned that they are far more diverse than we could have possibly imagined.
Some systems feature giant planets, larger than Jupiter, orbiting incredibly close to their star. Others host eccentric, solitary worlds, with no companions to call their own.
This wealth of data reveals one thing – planet formation and evolution is more complicated and diverse than we ever imagined.
As a result of these discoveries, astronomers developed two competing models for planet formation.
The first is core accretion, where planets form gradually, through collisions between grains of dust and ice. The theory has grown out of our old models of Solar system formation.
The competing theory is dynamical instability. Once again, the story begins with a disk of material around a youthful star. But that disk is more massive, and becomes unstable under its own self-gravity, causing clumps to grow. These clumps rapidly form planets, in thousands of years.
Both models can explain some, but not all, of the newly discovered planets. Depending on the initial conditions around the star, it seems that both processes can occur.
Each theory offers potential to explain eccentric worlds in somewhat different ways.
In the dynamical instability model you can easily get several clumps forming and interacting, slinging one another around until their orbits are both tilted and eccentric.
Under the core accretion model things are a bit harder, as this method naturally creates co-planar, ordered planetary systems. But over time those systems can become unstable.
One possible outcome is for one planet to eject the others through a series of chaotic encounters. That would naturally leave it as a solitary body, following a highly elongated orbit.
But there is another option. Many stars in our galaxy are binary – they have stellar companions. The interactions between a planet and its host star’s sibling could readily stir it up and eventually eject it, or place it on an extreme orbit.
This brings us to our newly discovered world, HD76920b. A handful of similarly eccentric worlds have been found before, but HD76920b is unique. It orbits an ancient star, more than two billion years older than the Sun.
The orbit HD76920b is following is not tenable in the long-term. As it swings close to its host star, it will experience dramatic tides.
A gaseous planet, HD76920b will change shape as it swings past its star, stretched by its enormous gravity. Those tides will be far greater than any we experience on Earth.
That tidal interaction will act over time to circularise the planet’s orbit. The point of closest approach to the star will remain unchanged, but the most distant point will gradually be dragged closer in, driving the orbit towards circularity.
All of this suggests that HD76920b cannot have occupied its current orbit since its birth. If that were the case, the orbit would have circularised aeons ago.
Perhaps what we’re seeing is evidence of a planetary system gone rogue. A system that once contained several planets on circular (or near circular) orbits.
Over time, those planets nudged one another around, eventually hitting a chaotic architecture as their star evolved. The result – chaos – with most planets scattered and flung to the depths of space leaving just one – HD76920b.
The truth is, we just don’t know – yet. As is always the case in astronomy, more observations are needed to truly understand the life story of this peculiar planet.
One thing we do know is the story is coming to a fiery end. In the next few million years, the star will swell, devouring its final planet. Then, HD76920b will be no more.
Jonti Horner, Vice Chancellor’s Senior Research Fellow, University of Southern Queensland; Jake Clark, PhD Student, University of Southern Queensland; Rob Wittenmyer, Associate Professor (Astrophysics), University of Southern Queensland, and Stephen Kane, Associate Professor, University of California, Riverside