Ep. 512: Direct Imaging of Exoplanets

Finding planets is old news, we now know of thousands and thousands of the places. But the terrible irony is that we can only see a fraction of the planets out there using the traditional methods of radial velocity and transits. But the new telescopes will take things to the next level and image planets directly.

In this episode we mentioned donations and tours. Click to learn more!

Download MP3| Download Raw Show with Q&A| Show Notes | Jump to Transcript or Download

Show Notes

Doppler shifting of planets and stars – red/blue shift due to motion
The Doppler Effect

CCD photography

Transit Photometry


CORAVEL, Elodie, Coralie, and HARPS – missions for planet detection

ESPRESSO (Echelle SPectrograph for Rocky Exoplanet- and Stable Spectroscopic Observations)[ – a third-generation, fiber fedcross-dispersed, echelle spectrograph mounted on the European Southern Observatory‘s Very Large Telescope(VLT).
Transiting Exoplanet Survey Satellite (TESS)


Fraser: Astronomy Cast, Episode 512: Direct Imaging of Exoplanets. Welcome to Astronomy Cast, your weekly facts-based journey through the cosmos where we help you understand not only what we know, but how we know what we know. I’m Fraser Cain, publisher of Universe Today. With me, as always, Dr. Pamela Gay, a senior scientist for the Planetary Science Institute and the director of CosmoQuest. Hey Pamela, how’re you doing?

Pamela: I am doing well, how are you doing?

Fraser: I’m doing great. First episode that we are recording in the New Year, 2019. Feels like only a marginal dumpster fire so far. Should be a good year.

Pamela: That’s optimistic.

Fraser: As we mentioned last week, there’s a ton of really fascinating space science, astronomical stuff that’s going to be happening this year. Everything else is irrelevant to us. All we care about is space, space, space. And that is not—

Pamela: Keep it simply science is the other keep it simple.

Fraser: Keep it—yeah, exactly, keep it simply space. Not even full science— even that is too—there’s too much chance of things going awry. Cool. Well, how did the fundraiser with CosmoQuest go, where are we at?

Pamela: So, by December 31st, we’re still waiting for the final numbers to come in, but it looks like we brought in $35,000. This is really impressive given that we started the fundraiser the day after the U.S. government shutdown, which means that there’s a whole lot of people out there who really care about our show who don’t know when their next paycheck is going to be.

It’s super impressive because the Dow Jones collapsed that day as well, so I’m really proud of how many people that in these super turbulent times said, “Helping this science project, helping people better—do, learn astronomy and space science matters to me, and I’m going to put my wallet where my mouth is.” And thank you to everyone who showed you want us to keep going and I just hope we can make you proud.

Fraser: Yeah. Yeah, thank you so much, everybody who contributed. Pamela’s got some runway to run CosmoQuest for the next few months, and hopefully we can get to a point that we’ve got more funds coming in from some of the grants and such, and we’ll just keep on doing science. Especially now that OSIRIS REx is at Bennu and that little spacecraft is gonna need some help choosing a rock.

Pamela: Any day now we’re going to be getting that first mosaic and I can’t wait.

Fraser: Yeah, that’s gonna be exciting. All right. Finding planets is old news, but we now know of thousands and thousands of them. But the terrible irony is that we can only see a fraction of the planets out there using the traditional methods of radio velocity and transits. But the new telescopes will take things to the next level and image the planets directly. All right, Pamela. So today we’re going to be talking about direct imaging planets. Before we get into direct imaging, let’s just go back a bit and talk about the traditional ways of finding planets orbiting other stars.

Pamela: Well, the—

Fraser: The old school way. The way our grandparents used to do it.

Pamela: The initial way that people found planets was they looked for the Doppler shifting of regular, everyday stars by hot Jupiters. And because the gravitational pull of a planet will, indeed, yank around its sun, we measured those motions the same way a police officer measures if you’re speeding or not— by seeing how light coming from the star could shift it red-wards or blue-wards in its motion. Now, this was good in 1995, ‘98, early 2000s. But it was quickly realized that you can actually see the dip in light from a star when a planet passes in front of it.

This was initially done mostly with stars that we knew had planets going around and around them, and after people with, like, four-inch driveway telescopes started being able to replicate planet results—well, we launched this little spacecraft that just might have been called Kepler, and we started pulling in planets a thousand at a time. Potential planets a thousand at a time. And since then, we’ve been going through confirming them using the Doppler technique, and sometimes even—well, lately we’ve found a tiny, tiny handful of planets by looking directly for them.

Fraser: All right. So, before you get into the direct imagining, you mentioned that amateurs with four-inch telescopes in their driveway can confirm exoplanets. So if you’ve got a small telescope and you’ve got, like, a CCD camera attached to your telescope, or you’ve got a DSLR attached to the telescope, then you can do this yourself. You can take the exoplanet catalogue that exists right now, you can find out when a planet is going to be transiting in front of its star, you can make observational measurements of the star over several nights and measure the amount of light that’s coming from the star.

And you can detect a noticeable dip in the amount of light that’s coming from that star. You can replicate, you can confirm, that there is a planet orbiting that star. And it’s amazing to me that we didn’t know this twenty, thirty, forty years ago. Like, this is a thing we could have done if we had sort of just known what to do. And I guess if the CCDs were good enough—

Pamela: They weren’t. This is the thing— this isn’t something you can do glass plate photometry. The time that it takes to expose the glass, measure the light, the errors in it— can’t do it that way. So you needed to have the invention of the CCD. The early CCDs didn’t have the precision, the quantum efficiency. We’re looking for tiny, tiny, tiny dips, sometimes just a tenth of a percent of light in the light from a star. And this is something we’ve only been able to do for the past couple of decades.

Fraser: All right, so let’s talk about the downside of both the radio velocity and the transit method, which is, of course, alignment.

Pamela: With the radio velocity method, we are limited to finding planets that are moving towards us and away from us in the plane of the sky. This means, ideally, we can only find things that, first of all, are big enough to yank their star around, and second of all, are close enough that they’re big enough has an effect. And you have to have the geometry where they’re moving in and out, towards and away from us in the plane of the sky. Okay, so that’s all annoying and limits what we’re able to see.

Now, with the transit method it gets even worse. With the transit method, we have to have the planet precisely aligned so it passes in front of its star, eclipsing some of the light from the star relative to us. Which means the alignment has to be not just in and out of the sky, but precisely so that it’s not tilted up or down too much. Even in our own solar system, we don’t see Venus and Mercury regularly eclipsing our sun. If we only catch Venus every few generations having a pair of transits, how much rarer is it going to be to see alien planets transiting their stars? Well, it turns out it’s pretty common, but it still limits what we can see.

Fraser: Yeah. And so what percentage of the planets out there actually line up so we can see them?

Pamela: Well, that would require me to know how many stars have planets. And so—

Fraser: Right. But assuming a star has a planet, what are the chances that it’s going to be lined up in a way that we can actually detect it using, say, even just the transit method?

Pamela: It’s only a few percent.

Fraser: Like, are we looking at, are we talking tens of percentage, or like a couple of percent, a half of a percent?

Pamela It’s a few percent. It depends on how far away it is–

Fraser: A few percent at the most.

Pamela: It’s even gonna be harder to find the ones that are further out, but yeah, it’s a few percent.

Fraser: Right. So for every one planet that we can see, there are fifty to a hundred planets that we can’t see because they just don’t happen to be lined up and everything is random. That sucks. All right, let’s get on to the direct imaging method then. What are we doing here?

Pamela: So with the direct imaging method, we are saying we don’t care what the star is doing, we don’t care about the alignment between the planet and the star. In fact, that star is our enemy. We’re gonna do everything we can to ignore its light, and strictly look directly at the light reflecting off of that planet or the light being emitted by the planet. This is often done in the infrared, where worlds like Jupiter sometimes give off more light than they take in from their sun. So we often use what are called coronagraphs to block out the light from the star.

This was most famously done with the star Fomalhaut, which any of you who are correctly located on the planet are able to go out and see. And with this little world, they blocked out the central star, and then imaged the dust disc around it, saw eddies in the dust disc, and used mathematics to predict where in that mass the planet should be located. And over a period of years, watched that planet systematically orbit around Fomalhaut. And it wasn’t for the fact that we’ve seen the sucker move, it just looks like another flicker in the dust disc. These things are really tiny because they’re so far away.

Fraser: And so this idea of this coronagraph—I mean, one does not just simply block the light from a star. Why is this so complicated?

Pamela: Well, it gets differently complicated with different systems. The primary issue is you have sunlight coming in from, or I guess starlight coming in, in a wave. You can’t just stand in one place on the planet and go, “okay, here I see the star,” and then step to the left and “here I can’t see the star.” No. That star is giving off light in a perfect sphere in all directions. Occasionally, gravity bends that sphere hither and yon, but in general, the starlight is coming in in all directions. So you have to figure out how to block all possible paths into the telescope for that light without blocking anything else.

Fraser: And how do they do that?

Pamela: Well, it depends on the telescope. My favorite example of a coronagraph is the one that is being developed potentially to fly with the WFIRST spacecraft. This is a telescope that it turns out the military had, essentially in a box waiting to be launched, and—

Fraser: Right. Yeah, you tell this story every now and then, right, this Hubble-class telescope that the Air Force used, or the National Reconnaissance Office would have used for Earth observation, but it wasn’t good enough anymore, so they were just, like, they said, “Hey NASA, do you want this, maybe? Otherwise we’re just gonna throw it out because it’s garbage for what we can do.”

Pamela: And the annoying thing is they had two and we could only afford one.

Fraser: I’ll take the other one.

Pamela: Yeah. Yeah. So we’re in the process of hopefully—assuming the government opens back up again—of building WFIRST. And one of its accompanying— I don’t know if the right word is instrument, facilities— the thing of it that’s hopefully going to get built to fly alongside it is a massive sunflower-shaped coronagraph that is essentially a solid disc surrounded by petals. These things come in a variety of different designs. They all look like something that would be out of a James Bond movie, used to kill somebody, throwing stars; whatever analogy you feel like.

It’s that combination of different edges that works to build interference patterns in just the right way to block out the star and allow you to potentially see whatever planets may be around the star.

Fraser: And these—I know this star shade is gonna fly, like, 10,000 kilometers away from the telescope and perfectly position itself to have the center part block the star, and then you’ll have the petals sort of gathering up some of that additional light that—that’s pouring out. And any planets, in theory, are going to be seen around it.

But even— I don’t know if you’ve, like, researched into coronagraphs much—there’s another way that’s really interesting, the way with some of the Earth-based ones and some of the space-based ones will work. The coronagraph is built into the spacecraft, or into the telescope, and uses destructive interference of the light so that it takes the light, breaks it up, makes the light fight itself, so that you actually, you get the block of—you still have a disc that blocks the sun. But then you also have, using destructive interference of the light to clean up all of it. And these will work best in space, but they still work okay in the atmosphere. And there’s a tremendous instrument that’s attached on to the European Southern Observatory. They have the Espresso—

Pamela: Very Large Telescope and Keck both have these instruments.

Fraser: Yeah. Yeah, these coronagraphs, I think Espresso is the one that’s on the VLT?

Pamela: I think so.

Fraser: Yeah. And so you were saying that these work better in infrared than visible?

Pamela: Infrared. And the issue that makes infrared better than visible light is that the wavelengths are just a little bit longer, and that difference in size of the wavelength of the light is, in a lot of ways, easier to deal with. Now, that’s not the only thing. The other thing is stars give off less light in the infrared, it’s below the peak in their relationship between color and how much light output they have. And it’s in the maximum light for a planet.

So planets are warm objects— they’re not hot, they’re not stars. So we’re starting to get an infrared towards the peak temperature that—or the peak wavelength corresponding to temperature that a planet gives off its light and we’re getting below the peak wavelength for a star. So this also helps. We’re just looking in a different color and we’re getting a myriad of bonus features. And sometimes that’s all it takes— is just those extra bonus features the universe gives you for free if you just shift out of what the eyeball can see.

Fraser: And one of the things, also, that the Espresso instrument is able to do on the Very Large Telescope is it’s able to detect the polarization of the light. And so you get the light coming off of the star, it’s gonna be bouncing off of the planet’s atmosphere, and possibly oceans and things, and it polarizes light and changes it into a different alignment. And then the instrument can detect, can throw away everything that is not that exact kind of polarized light, so that you can only see the movement of that planet. Which is essentially a tiny little polarization spot, moving around. And I know earlier this summer, again, the European Southern Observatory detected a newly forming planet for the first time using this technique.

Pamela: And the thing that is so frustrating is that many of the objects that we’re saying are direct detections of planets are being found around brown dwarfs, are being found around red dwarfs, are not necessarily planets themselves because they might be brown dwarfs. So many of these discoveries, many of these current direct detections have asterisks next to them, just because we’re not entirely sure which side of the planet’s star line we’re on. And, to remind everyone, a brown dwarf is a object that is massive enough that it may have been able to burn tritium in its core, but it only had temporary nuclear fusion going on inside the star.

And it’s that temporary nature that it doesn’t actually start hydrogen burning that leads us to—it’s not a planet, not a star, it’s its own brown dwarf kind of thing. Now, do you call something that’s orbiting a brown dwarf a planet? Sure, why not. But do you call it a direct detection of an exoplanet, or do you call it the direct detection of a rock around a thing that isn’t a star? So you put an asterisk. And then there’s all of the stars that have many Jupiter-mass objects being detected at fairly significant distances away from the star and the question is, are these things that are super-Jupiters, are these things that are brown dwarfs, what the heck do you call them? With our present-day direct detections, there’s an asterisk left and right.

Fraser: So here we are now—I mean, I think we’re in the phase where people are producing these first images, they’re showing what is possible with the low-hanging fruit, of Fomalhaut, which is the star and the way the planet work, it’s sort of the perfect contrast. What is in the works then, to take this whole plan to the next level? You mentioned the star shade, which is one idea. What are some of the other—what’s in the works that’s going to allow us to take this to the next level?

Pamela: Well, I think that what was done with Fomalhaut is potentially going to start giving us a new way to look for planets. This is research that matches in beautifully with a lot of what we see coming out of the Atacama Large Millimeter Array, where, day after day, we’re seeing new images of proto-planetary discs with planets in the process of forming. But, in these radio wavelengths, we aren’t able to see the planets. By coupling the Atacama Large Millimeter Array observations with future VLT ground-based observations that allow us to take advantage of the infrared, take advantage of the technologies like Espresso, we’re going to have a new way to find specific kinds of planets that have already shown us where they are by how they gravitationally move around the discs they’re in.

Now, finding the isolated planets, that’s gonna be more complicated, and this is where we have to keep looking to the future. A spacecraft that— maybe, maybe, maybe, please give us Terrestrial Planet, find her, we’ve been begging for thirteen years now— we really are going to have to wait for the space-based missions like WFIRST to get off the ground, both figuratively and literally.

Fraser: Right, WFIRST— I mean, James Webb would probably be able to do it if it was connected with the star shade. The great thing about this idea of the star shade is it can work from multiple spacecrafts. Many can use it. But the downside is that you’ve gotta get this alignment at 10,000 kilometers. You’ve got the spacecraft, you’ve got the star shade, you only get to look at what is perfectly behind the star shade from the position of the spacecraft. Get multiple spacecraft in now, they’re all using the same star shade, but the way the lines are coming, they’re looking at different targets. So that’s effective. I did a video— oh, go on—

Pamela: And you’re limited with the star shade and how long you can use them before. Because they do have this— in a single observation, because they do have this great separation between them, which means that they’re gonna be orbiting at very different velocities. And you’re limited on the diversity of orbits your telescopes can be in, because you have to be able to get into similar enough orbits. So a star shade designed for WFIRST orbiting the Earth in whatever orbit they put it in won’t work for James Webb space telescopes. So while you can share things, you can only share so far, and— I’ve apparently got to capture a puppy, I’m so sorry, everyone.

Fraser: Yeah, so one of things we’ve mentioned in the past is that there’s a whole new—there’s a few technologies in addition, like adaptive optics, which allow these ground-based telescopes to take even better pictures now than the best space-based pictures that we’ve seen. And you mentioned that idea of the very large array— there was just this amazing picture that came out about maybe two weeks ago, and there was a bunch, they had like twenty different proto-planetary discs captured by the VLA. And you see them in all different, face-on—

Pamela: That was ALMA.

Fraser: ALMA. Yeah, yeah. Face-on, a little off to the side, right? All the different variations you would want to see in other star systems. Now you’re seeing them at every angle, and so—

Pamela: And every stage in their formation, from being a fairly solid disc to having progressively more and more complex planet systems around them. We’re slowly building up a film, one system at one age at a time, of what it looks like for a planetary system to grow up. And this is truly amazing. It’s going to be finding the senior solar systems that no longer have those massive discs around them to gravitationally point us exactly to where around the star to look for that faint little planet. That’s where we’re gonna have our troubles.

Fraser: Yeah. But the next class of telescopes that are in the works, across the board, this is gonna be the kind of thing they’re attempting to do. You’ve got the Extremely Large telescope, which is the one that’s being developed for the European Southern Observatory. This is a 39-meter telescope, when you compare it to the four 8.4-meter telescopes with the Very Large Telescope. You’ve got the Magellan Telescope, you’ve got the 30-meter telescope, which is 30 meters, and then you you’ve got some of the space-based telescopes, James Webb, and then what comes after that? And this direct imaging of exoplanets is gonna be one of the capabilities that’s being baked into all of them. That, if this is the way, once we’ve got things like TESS and CoRoT and Kepler and CHEOPS finding all of the nearby transiting planets, then it’s time to move to a system that lets us see them all.

Pamela: And with Gaia and its extraordinarily precise astrometry, we have yet one more way to find planets, and that’s by seeing how they yank around their stars in the plane of the sky. So it’s not that the star is coming towards us and away and we can measure the Doppler shifts, it’s that it’s actually moving around relative to other objects, and—

Fraser: Right. Spiraling. And you can see the star doing a little wave in the sky as its planet is pulling it around.

Pamela: And this is something that is entirely new, something that we’ve been trying to do, hoping to do with objects like Barnard’s star for decades. But we’re finally able to do it thanks to the amazing high-resolution, high-quantum efficiency systems we have on orbit now.

Fraser: Yeah. And so that will serve as a great finderscope for these follow-on direct observations. Because you’re gonna know all of the nearby planets that are going to be easy to observe thanks to the Gaia data. So they all just build on top of each other. But it’s amazing to think that in, say, another 10 years or so, 15 years, that the main way that we look at planets now is gonna be doing these direct observations. We will have pulled away all the low-hanging fruit of the radial velocity and the transits, and then we’re just gonna start taking pictures of planets. What will it take, though, to give us that picture of another planet, like to see mountains and oceans and continents and things like that? What’s that gonna take?

Pamela: A generation starship and waiting a few years for the signal to get back once they get there?

Fraser: Right. So we’re gonna need to send breakthrough star shots off to other star systems before we can actually see what they’re gonna look like up close. The one idea is this idea of using the light from the sun as a gra— using the gravity of the sun as a gravitational lens. The only challenge you gotta do is get your spacecraft out to about a thousand astronomical units from the sun, and then you can take a picture of other star systems. But then it’s amazing. Like, you get to see them— then you do get to see mountains or whatever happens to be perfectly lined up behind the sun.

Pamela: And it’s that trick of getting it perfectly lined up, because if you aren’t perfectly lined up, you end up with a streak, a smear, a smiley face. There’s so many shapes you can get that aren’t perfect. So, I don’t like funhouse mirror worlds. Let’s just go there instead.

Fraser: Oh, I say we just do both.

Pamela: Okay.

Fraser: All right. Thanks, Pamela. Thanks, everyone. We’ll talk to you next week.

Voiceover: Thank you for listening to Astronomy Cast, a nonprofit resource provided by the Planetary Science Institute, Fraser Cain, and Dr. Pamela Gay. You can find show notes and transcripts for every episode at Astronomy Cast. You can email us at info@astronomycast.com, tweet us @AstronomyCast, like us on Facebook, and watch us on YouTube. We record our show live on YouTube every Friday at 3 p.m. Eastern, 12 p.m. Pacific, or 1900 UTC. Our intro music was provided by David Joseph Wesley, the outro music is by Travis Searle, and the show was edited by Susie Murph.

[End of Audio]

Duration: 29 minutes

Download MP3| Download Raw Show with Q&A| Show Notes | Jump to Transcript or Download

Follow along and learn more: