Alex Teachey speaks to SciTech Europa about the potential discovery of an exomoon outside our own Solar System.
Using the NASA/ESA Hubble Space Telescope and older data from the Kepler Space Telescope, astronomers have found the first compelling evidence for an exomoon outside our own Solar System. The data indicate an exomoon the size of Neptune, in a stellar system 8000 light-years from Earth.
SciTech Europa asked Alex Teachey, an NSF Graduate Research Fellow at Columbia University’s Department of Astronomy in the USA, who was involved in the research, about this potential exomoon candidate, the instruments used, and his hopes for the future of hunting for exomoon hunting.
What are the biggest challenges involved in detecting an exomoon, as compared to exoplanets, for example (i.e. transit, moon shift with each transit etc.)?
In general, we expect that an exomoon will be small, typically the size of the Earth or smaller, based on what we see in our Solar System and on theoretical modelling. Small planets are exceptionally difficult to detect because they block out very little starlight as they transit. To find these small objects we therefore typically need a combination of favourable circumstances: we want the star to be small and nearby, and we want the planet to have a short orbital period so we can see it transit many times. All of these will improve the quality of the signal.
With exomoons, however, the difficulty is compounded because, unlike planets that transit more or less like clockwork, a moon is expected to show up in a different place every time the host planet transits, sometimes before the planet, sometimes after, and this means it’s not so easy to stack transits in such a way that will improve the signal. In short, they tend to be ‘lost in the noise’.
On top of that, we then have to build a self-consistent model of the planet-moon system that not only explains what we are seeing across every transit event, but which does so much better than the planet-only model so that the moon hypothesis is convincing enough to make a claim.
Finally, the picture is now emerging that an exomoon is more likely to be found orbiting colder planets, farther from their host stars, and this poses all kinds of new challenges; we have fewer transits to work with, and the probability of seeing these planets’ transits at all falls off dramatically as we get further from the star.
How important are instruments such as Kepler and the Hubble Space Telescope (HST) to your work?
We initially identified an exomoon candidate orbiting Kepler-1625b in the Kepler data, and that has been a phenomenal mission and proving ground for the search for an exomoon. Unfortunately, even with the incredible capabilities of Kepler, we are not able to probe these colder regions of space (i.e. further from the star) in the kind of detail we would like, because the primary mission ended prematurely.
Hubble was clearly a very important part of the present analysis, with about a factor of four improved sensitivity over Kepler. But targeted observations like the one we performed are limited time-wise so they cannot give us a lot of ‘out of transit’ data to work with, and that is very important because that is where we expect to find moons, and because that data is valuable for characterising what is going on with the star.
Targeted observations for the follow-up of candidate exomoon signals is going to be very important going forward, but what would be ideal would be a spacecraft like Kepler, with improvements in sensitivity but also, critically, a longer observation baseline (up to 10 years or more). That is going to be crucial for finding more of these cold transiting planets.
In the potential discovery of a moon orbiting Kepler-1625b, there is the suggestion that the planet’s wobble could be caused by the gravitational pull of a hypothetical second planet in the system, rather than a moon. How important a hypothesis is this? How could it be confirmed?
The presence of another planet in the system is certainly plausible, and we can’t rule it out right now. To be clear, no other planet has been detected in the system, but it could be there. The exomoon hypothesis has the added benefit that it explains two observations, namely, the transit timing variations and the apparent flux dip at the end of our Hubble observation, while the second planet hypothesis only really explains the timing variations adequately. (The question has also arisen whether the flux dip could be another transiting planet, but we calculate a very low probability for that being the case).
In any case, follow-up radial velocity measurements of the system should be able to establish whether there is indeed another large planet in the system, and we are hopeful that we may perform another observation of the planet transit in May 2019, which is really what is needed to confirm or refute the exomoon hypothesis.
You have reported that the candidate moon is unusually large – potentially comparable to Neptune. What role could this play in developing new insights into the development of planetary systems and could it mean that theories of how moons form around planets will need to be revised?
The first question everyone asked us was, “how do you get something like this?” The existence of a satellite this size has hardly been anticipated, precisely because it is not easy to see how you get it through known mechanisms, of which there are essentially three:
- A collision such as that which we think gave rise to Earth’s moon;
- A capture scenario, similar to what we think happened with Neptune’s moon, Triton; and
- Formation in a disk of material orbiting the planet in the early years of the system’s formation.
Personally, I wouldn’t rule out some version of any of these scenarios to explain this exomoon, if it exists, but I am not a dynamicist. If it is any of these three scenarios, there are some new questions to answer.
The nice thing about being an observer is that sometimes we find things no one expects or has an explanation for, but we find out that Nature has no problem producing them. To highlight just two of my favourite examples: we know there are objects called ‘Hot Jupiters’, which are very large planets orbiting extremely close to their host star, but more than 20 years after the first discovery of a Hot Jupiter the theorists are still debating how such a thing can form. And, closer to home, the theorists still haven’t got a great answer for how you get from a protoplanetary disk made of microscopic particles and build planets out of that. That is, how did our Solar System come to be? Nature has no problem making planets, clearly, but we are still trying to figure it out.
How do you hope the James Webb Space Telescope (JWST) will aid in the discovery of exomoons (and indeed exoplanets), and where will your research priorities lie in the future?
JWST will be a fantastic instrument when it launches, and I certainly hope we’ll be able to use it for targeted exomoon observations. It will be more sensitive than Hubble, so we could see these objects better than ever before. Of course, everyone’s going to want to use it, so it will be extremely ‘oversubscribed’, and getting time will be tough. And for the exomoon search we need a lot of time. All this is to say, we’ll definitely be looking at using JWST, but I really hope HST will continue to operate for many years to come.
For the aforementioned reasons, we are turning our focus more towards the cold planets, and there are all kinds of new challenges associated with going after those targets. There is lots more work to do, but for us that is a good thing.