The first attempted launch of TESS on Monday afternoon was called off for some additional tests of the guidance system. But the Falcon 9 is ready on the pad for a second attempt at 6:51 PM ET.
Unfortunately, I’m not on site for this one. But while I was at NASA’s press office on Monday, I had the opportunity to talk with Doug Hudgins. Dr. Hudgins is the program scientist for NASA’s exoplanet missions and deputy program scientist for TESS. We hid away in an small office behind the information desk to discuss what we’ve learned from the Kepler and K2 missions, and what we expect to learn from TESS.
DK: When we look out there, we don’t seem to see systems that look like our own. To what extent is that a bias imposed by the instruments we’re using?
Hudgins: In my opinion, I think it’s entirely bias. If you look at the distribution of planet sizes, you have a smaller number of the very largest planets, and it ramps up to a high abundance at around the Neptune size, about four times the size of the Earth, and then the distribution is kind of flat all the way down to what we can usually find, planets about one to one and half times the size of the Earth.
The evidence is that rocky planets are common. The problem with finding planets like ours is just that we’re small, and relatively far away from our star. So we’re hard to detect whether you’re looking at transits or the wobble of the star. For Kepler, the Earth is one of the smallest planets it was designed to detect, and with only three transits or less, it could have been missed in the noise.
Come on in to read the rest of the interview, and get ready for launch.
Thursday, Apr 19, 2018 · 2:45:32 PM +00:00
·
Mark Sumner
The second time proved the charm. TESS successfully launched on Wednesday afternoon at 6:51 PM ET. But there’s still work to do before TESS is in it’s unique orbit and ready to go to work.
NBC has some nice coverage of the launch with some background on TESS.
DK: What would we have seen of our system if it were 50 or 100 light years away, and we were looking at it with Kepler?
Hudgins: If Kepler had kept running for eight years, the estimate is that we may well have been able to see us. If we looked long enough, we could have seen an Earth orbiting a sun like ours. It’s all in the number of transits that you can see. Kepler found plenty of Earth-size planets. The problem was they were all orbiting very close to their stars. The reason we could find those, and not the ones that were out at the distance of the Earth, is because Kepler measured many, many transits. And because you had many, many transits, it could beat the noise down to see these tiny little signals. Since Kepler failed after just over four years, an Earth like ours would make three, maybe four transits—if you’re very, very lucky. And that might not be enough to detect it against the background noise.
DK: So even if you were looking for large planets—if you were looking at our system, and you were looking for Jupiter—you probably wouldn’t have seen it because it wouldn’t have transited at all?
Hudgins: Jupiter might have transited once during that period. We might well have seen Mercury, even though it’s smaller, because it would have made more transits. The challenge of getting to see an Earth is a big one, but everything we learned from Kepler shows that Earth-sized planets are out there, and abundant. It’s just looking long enough to see Earth sized planets in Earth-sized orbits.
DK: We used to have a very staid view of how planetary systems form. Rocky planets here. Gas giants over there. But when we look out there, it seems like there are a lot of things not in the right place. Do you think that’s because there’s a lot of swapping and bumping going on, or are there problems with our basic theory of planetary formation?
Hudgins: I think it’s pretty clear from the mind-boggling diversity of planetary systems that we’ve seen with Kepler and K2, and we’ll see with TESS, that planetary formation is a lot messier than our canonical ideas. It all made perfect sense—with one example—then exoplanets came along and ruined the story. We now understand that there are probably at least two different processes at work in planetary formation. One is the kind of core accretion model we’ve always talked about. Two bits of dust stick together. And more sticks to that. And you get bigger objects, and planetesimals, and so on. But there’s also what’s called gravitational instability. The same way that a star may initially form because of a slightly more dense area within a cloud of gas, planets like Jupiter probably formed in the same way. They’re essentially a mini-star that forms the same way the star did in the disk around the star. They just don’t get enough material to become stars—otherwise we’d have a binary system.
The other thing that affects the kind of planets we’re seeing that we probably didn’t appreciate before is timing. When those planets are forming by accretion, they may create rocky planets. But if they get started while there is still a lot of hydrogen and helium and other gases around, you end up with something like Uranus or Neptune—a rocky core with a lot of gas. On the other hand, by the time that Earth and Mars and Venus built up to the size where they could attract gases, where wasn’t a lot of gas left in the disk.
DK: One of the things that we’ve seen with Kepler that we didn’t expect is these water worlds. We’ve always thought about water as something that’s needed to create a planet like Earth, but now we’re seeing worlds that have too much water. That may be all water. How common do you think these things are going to turn out to be?
Hudgins: That’s a really interesting question. The answer is, I don’t know. I have a hard time getting my head around how you make a water world at all. I wouldn’t even hazard a guess. This is the fascinating thing about the whole science. Theoreticians are furious—the old model, it seemed so neat and straightforward, and now everything we find is breaking those models and making them create new ones.
DK: That seems to have happened in our own Solar System. We just knew what all those moons and outer planets were going to be like, until we got out there.
Hudgins: And we got there, and they were all different. This is what we explore. Because we’re always surprised. TESS will surprise us. We think we know what TESS is going to do, but TESS will find things that Kepler or K2 didn’t see.
DK: Is there something special you’re hoping to see? We all want to see Earth-like planets, but is there something else you’d love to see? Something we haven’t see that you think we might find?
Hudgins: Not really. What I’d like to see is close planets. That the closest stars to us have planets. I’d like to be able to take the kids out in the backyard and point at a bright star and say “We know there are planets there.” And that sets us up for a follow on mission that can look at the light from those planets and tell us what the atmosphere is made of, and maybe even something about its surface.
DK: Will the Webb telescope do that? Will it let us follow up on TESS and look at those atmospheres?
Hudgins: Yes, with an asterex. Webb will certainly follow up on the best TESS candidates.
DK: If Earth were 10 light years away, would Webb be able to image it?
Hudgins: It would be really tough. Because we’re not warm. Webb is an infrared telescope, it’s looking at heat. And Earth just isn’t that hot. The reflected light from Earth in the infrared might not be enough, even for Webb. Hubble and Spitzer have been able to look at a half-dozen or so hot Jupiters, large planets close to their suns and tell us something about their atmospheres. Webb will be able to look at smaller hot planets, and more distant large planets. We’ll be able to look at some planets that are not being absolutely roasted because they’re right on top of their star. A lot of these planets have hazes that probably come from minerals that are just being vaporized. It’s bizarre, but not what you expect to be the atmosphere of an average planet.
DK: Our system all seems to be still in order. But these hot Jupiters, it seems clear that these planets didn’t form where we see them today.
Hudgins: That’s right. The first planets we found were hot Jupiters, because they’re the easiest to find. This immediately blew up the traditional model. Theoreticians thought about it for awhile and said “Ah ha! They form at a distance and migrate inward.” And then they predicted that if you had one of these giant planets migrating through the system, it would wreck the system. So where you find one of these hot Jupiters, you wouldn’t find any other planets outside their orbits. And then we did. Now they have to figure this out. If you have planet migration, and it seems that you do, because it doesn’t seem possible for these gas giants to form near a star, how can planets migrate without clearing all the orbits between the star and where they form? That dance must be more elegant than a bull and a China shop that just smashes into or throws out everything in its path.
DK: How many systems out there are simply not inclined correctly so that we see the planets in transit?
Hudgins: Many. What’s really amazing is that we’re seeing systems with three, four, even five or more planets that transit their star. If you were looking at our solar system from more than a few light years away, you probably couldn’t find an angle from which you could see more than one or two planets in our system. There simply too much difference in the inclination of the orbits.
DK: So even if you were looking at our system from a distance, and you looked for a long time, you might see … Mercury and Jupiter. Or Venus and Earth. But not the rest of the system? We’re used to seeing pictures that make our planets all seem to be orbiting on the same plane.
Hudgins: But they’re not.
DK: So some of those systems have to be very flat, if we’re seeing so many planets.
Hudgins: Extremely flat.
DK: Is there any reason to believe that the orientation of these systems fits any larger kind of pattern, or are they random?
Hudgins: We assume that the orientation of orbits is completely random. And so far the data seems to bear that out. So it just breaks down to a geometrical argument. A planet at this distance from a star this size will only transit if it orbits within these few degrees. What’s amazing is how many we see, because there’s so many more that we don’t. The further a planet is from a star, that smaller that angle is. Which is another reasons we see so many planets that are very close to their stars.
DK: The Zooniverse site has offered up Kepler and K2 data through it’s Exoplanet Explorer program. Looking at transit curves seems to be one of those things were a “calibrated eyeball” can contribute and lots of volunteers like feeling that they’re involved in this search. Will the TESS data also be available?
Hudgins: Certainly the data from TESS will be incorporated into Zooniverse and you’ll be able to do the same thing. In fact, the same thing cubed, actually. The TESS mission, like the Kepler mission, is going to have some specific stars that will be monitored. But what I hazard will be the most exciting part of the mission overall, is that because of the orbit that will bring TESS close to the Earth every 14 days, we can get very high bandwidth communications with it. We’ll actually be able to download very detailed, full-frame images rather than the smaller images and lower resolution data from Kepler. So you’ll have not only a lot more stars to look at, you’ll have much better data.
Because of the full-frame images, there will be a mind boggling number of stars that need to be evaluated, and there are not enough professional astronomers in the world to do it. We’ll develop automated techniques to find the ones that are easy to find.