Another exoplanet joins the HR 8799 family

By Phil Plait | December 27, 2010 7:00 am

In late 2008, astronomers announced the discovery of a multi-planet system orbiting the star HR 8799. The three planets were discovered the old-fashioned way: they were directly imaged! [A gallery of all known directly-imaged exoplanets is at the bottom of this post, in fact.] The star is young (30 – 60 million years old), so the three planets are also young, and still glow with the leftover heat of their formation. In the infrared, they’re bright enough to be distinguished from their star in images.

And now, follow-up observations using the monster Keck 10-meter infrared telescope have revealed a fourth planet: HR 8799 e:

See it there? The star is the weird blob in the middle, with most of its light removed using techniques that allow the much fainter planets to be seen. The other three planets, labeled b, c, and d (the letter "a" is reserved for the star itself, and is usually just assumed) are obvious as well.

Some aspects of the planet are pretty easy to observe; given the distance (130 light years) to the star, we can measure the planet’s orbital size off the image; it’s about 2 billion kilometers out, a little bit closer in than the distance of Uranus from the Sun. It takes roughly 50 years for it to complete one orbit.

The most important characteristic of the planet is its mass, and that’s not all that well known. With some planets we can measure the gravitational tug of the planet on the star and deduce its mass from that, but that won’t work here (the orbital period is too long, and it’s easier if the plane of the orbit is edge-on to us, instead of the nearly face-on orbits of HR 8799′s planets). So we have to rely on models using the physics of how a planet cools after it forms; how bright it is now depends on its mass and its age. More massive planets glow more brightly than lighter-weight ones, and they fade as they age and cool. The problem is, we don’t know the age of the star well enough to nail down the planet’s mass; if it’s 30 million years old the planet is 5-10 times the mass of Jupiter; if it’s 60 million years old it’s 7 – 13 Jupiter masses. A good guess is probably about 7 – 10 times Jupiter. That’s pretty firmly in the planetary mass range; if it were 13 or more it would be called a brown dwarf.

The thing is, it’s not clear a planet of that mass can form that close to the star. We don’t know everything there is to know about how planets form, but the current thinking is that something like this can form far more easily farther out, and then over millions of years it migrates closer to the star. There are a couple of ways this can happen; for example, there can still be lots of dust and junk orbiting the star leftover from its formation, and as the planet plows through this stuff it loses energy and gradually falls toward to the star. Interactions with other planets can also move it closer to the star as well.

The cool thing is, cases like HR 8799 e help us test our models, showing us where the holes are and making them better. And since HR 8799 is now known to host a full-blown solar system full of massive planets that can tug on each other, it’s a fantastic testing ground for our understanding of the complex physics involved with making planets. And even better: over time we can watch as these planets physically move around their parent star, and we’ll get even better data to feed our models. It’s pretty amazing that we can see planets orbiting other stars at all, let alone learn so much from them.

Tip o’ the deuterium limit to Universe Today.



[Below is a gallery of exoplanets that have been directly imaged using telescopes on ground and in space. Click the thumbnail picture to get a bigger picture and more information, and scroll through the gallery using the left and right arrows.]



CATEGORIZED UNDER: Astronomy, Cool stuff, Pretty pictures
MORE ABOUT: exoplanets, HR 8799

Comments (28)

Links to this Post

  1. Kort astronieuws van de afgelopen week | Astroblogs | January 2, 2011
  1. breadbox

    Another win for the ground-based telescopes!

  2. Mike Saunders

    @breadbox
    Have to say I found a few transiting planets from kepler data just even today ;)

  3. Why the lateness of this post? We discussed this at the astro-ph coffee on 23 Nov 2010, five weeks ago. I suppose it is some Nature embargo issue. So read arXiv.org

  4. Nate

    Hey Phil, did you forget planet e? Without reading the paper “Images of a fourth planet orbiting HR 8799″, it looks like four, not three. :)

  5. What is the orientation of this system relative to us? Are we looking “down” on the orbits? Do we know the angle?

  6. Ned (4): I know the news came out a while ago; I happened to see this on Universe Today a while back. I made a note to myself, and finally got to writing it up. I have lots of notes like that, but limited time to get to everything.

    Nate (5): I don’t understand what you’re saying. There are four planets (at least) orbiting HR 8799. That’s the point of this post. Am I misunderstanding your comment?

  7. Wow, four planets in one picture! There is always something a little special about directly imaging planets. Doubly interesting when things seem to break our current model of how planets and solar systems work!

  8. Nate

    No, I can’t read is all. It’s early for me. ;)

  9. andy

    That’s pretty firmly in the planetary mass range; if it were 13 or more it would be called a brown dwarf.

    I wish people would stop quoting the 13 Jupiter masses limit as if it was gospel. For starters, the onset of deuterium fusion in an object’s interior (which is where the 13 Jupiter masses number comes from) is dependent on various factors that can alter this number quite significantly.

    Furthermore, even if the mass of this object exceeded the deuterium fusion limit it would constitute yet more evidence to an already fairly convincing case that using the deuterium fusion criterion is a nonsensical way of dividing planets and brown dwarfs that only leads to confusion and unnecessarily complex descriptions of real planetary systems.

    Taking the example of a different but well-known system, the second planet out from Upsilon Andromedae A has a mass exceeding 13 Jupiter masses. However it is clearly part of a system of planets (there are three other gas giants known in the system) – calling one of them a brown dwarf because it is above the deuterium fusion threshold makes the description of this system needlessly complex.

    There might be a case for making a mass cutoff, but the observational evidence suggests the boundary probably lies at around 30 Jupiter masses for objects orbiting solar-type stars, and that is above the deuterium threshold. Following the evidence rather than gut feeling leads to the conclusion that in this universe, there are planets which undergo deuterium fusion in their interiors in the early stages of their lifetimes.

  10. Mike Saunders

    Hey Phil,

    How are these images created? They don’t look exactly like “Direct” images, but instead like some algorithm has been run over the data using complex information, like MUSIC or something. The artifacts around the star suggestion some sort of Fourier transform has been done at some point, I’ve seen similar things in my own work. I’d imagine that otherwise the information about the planet would be buried deep within the star noise without some sort of correlation or comparison method.

  11. Yorin

    Hi Phil,

    your calculation of the planet’s orbital size relies on the fact that we see the orbit face-on. I can’t see any movement on these pictures (which is no surprise since the orbits are so far out). So how do we know that?

  12. Gary Ansorge

    Planets moving from a high orbit inward seems to be the best explanation of planetary formation such as we see around Sol, as in, rocky bodies left behind as their gases are boiled off by their sun.

    Great pics, Phil.

    Gary 7

  13. Messier Tidy Upper

    Wonderful news – what a great Christmas / Hanukkah / Sol Invictus / [insert festival of choice here] present from the cosmos and the Keck telescope discovering scientists. Thanks as always to them and the BA too. Nice write-up & blog post here as well. :-)

    I’ve got just one very minor nit of sorts :

    The problem is, we don’t know the age of the star well enough to nail down the planet’s mass; if it’s 30 million years old the planet is 5-10 times the mass of Jupiter; if it’s 60 million years old it’s 7 – 13 Jupiter masses. A good guess is probably about 7 – 10 times Jupiter. That’s pretty firmly in the planetary mass range; if it were 13 or more it would be called a brown dwarf.

    Hmm .. Given the uncertainties in the mass range there, the fact that thees are minimum mass estimates (or so I gather) & the fact that that’s at the high end of mass even for superjovian exoplanets I’m not so sure “pretty firmly” is the best choice of words here.

    Then again could we have a system where a star is orbited by a set of brwon dwarfs not exoplanets? Do they two categories overlap to some extent? Seems to me this area is stuill failry fuzzy with many semantic points and competing definitions here leading to lots of questions raised but not clearly answered.

    Whatever the case and however these gaseous spheres are defined though, I consider this HR 8799 or “Gadolabovese”* system to be a particularly fascinating heavyweight discovery in the exoplanetary field. :-D

    ————

    * As I like to call it – derived from the star’s unique trifecta of being a Gamma Doradus (very slight & subtle) type variable and also a Lambda Bootis metal poor type star with a Vegan style circumstellar dust -and protoplanetary disk. :-)

  14. Messier Tidy Upper

    As is now almost traditional, I’m going to post this link via Kaler’s superb Stars website for folks :

    http://stars.astro.illinois.edu/sow/hr8799.html

    giving y’all more info and a photographic finder chart for Gadolabove. (HR 8799.) Interestingly, it is located directly opposite the sun of the first exoplanet discovered around a sun-like star, the Hot Jupiter “Bellerophon” or 51 Pegasi b, and is just inside the prominent sky-mark (as opposed to landmark) the Great Square of Pegasus.

    More about Bellerophon via wikipedia here :

    http://en.wikipedia.org/wiki/51_Pegasi_b

    & the newest (?) exoplanet discovery here :

    http://stars.astro.illinois.edu/sow/nuoct.html

    via Jim Kaler’s planet project page. :-)

  15. Messier Tidy Upper

    @3. Mike Saunders Says:

    @breadbox – Have to say I found a few transiting planets from kepler data just even today

    Congratulations! Superb news to hear – well done. :-)

    Can you tell us any more about that please? Or are you not allowed to?

    Finding news worlds around other stars. I love it. :-)
    [Envious] Sigh. Wish I had your job! [/envious]

    @6. kuhnigget :

    What is the orientation of this system relative to us? Are we looking “down” on the orbits? Do we know the angle?

    I’m not 100% sure and please correct me if I’m wrong, but I gather it seems to be face on so we’re looking at the Gadolabove star and its retinue of superjovian (proto?)exoplanets / brown dwarfs from directly above / below.

    *****

    PS. Sorry folks about all the typos in my earlier comment(s) – afraid I keep seeing the words that I *think* I’ve typed rather than the one’s I actually have. Sigh. :-(

  16. Joseph G

    Simply amazing! I didn’t know that many exoplanets had been imaged. And wonderful image gallery – I like the “timeline” descriptions.

    Just curious – the light cancellation technique Phil describes, does it have a name we can Google?
    I’d like to know more about how the starlight is masked out – are they using physical occulting disks or grids, some kind of optical wizardry using filters and prisms, or after-the-fact computer processing of the data? Or all three?
    Also, could this technology be adapted to search for asteroids that orbit very close to our sun?

  17. ad

    Ok, a system with 4 planets (so far, there might be more in close). Should be good enough to test for a Titius-Bode Law scenario.

  18. Messier (17).

    Transiting planets from Kepler data. I’m pretty sure Mike Saunders is talking about this: http://www.planethunters.org/

    You can do it, too.

    Andy (10) – totally.

  19. Jon Hanford

    @18 Joesph G

    “Just curious – the light cancellation technique Phil describes, does it have a name we can Google?
    I’d like to know more about how the starlight is masked out…”

    According to their paper, a technique known as Angular Differential Imaging was used to acquire these images. Here’s a non-technical description of ADI: http://www.mpia.de/homes/thalmann/adi.htm

    (there’s also links here to more technical info and a 2009 study of the GJ 758 system employing this technique)

    Pretty cool, eh!

  20. Joseph G

    @21JohnHanford: Thanks much!

  21. baric

    Am I imagining things or is there a 5th circle in that last image, halfway between planet e and the star? It’s white and seemingly circular, in between the black band and the red dust area.

  22. Messier Tidy Upper

    @ ^ baric : Honestly, I’m not sure. That could be something or could just be part of the star itself or an image artefact. I would presume the astronomers involved would have checked everything out closely & ruled it as non-planetary in nature but then I don’t know that for sure & could be wrong.

    @20. Vagueofgodalming : Thanks, that does indeed sound most probable. :-)

  23. Isaac

    I’m confused. At one point, Dr Plait writes, “… it’s about 2 billion kilometers out, a little bit closer in than the distance of Uranus from the Sun.” Later, he writes, “The thing is, it’s not clear a planet of that mass can form that close to the star.”

    The orbit of Uranus is close to the Sun? Just how far out are large planets expected to form?

  24. andy

    @Isaac: IIRC the prime region for forming gas giants around solar-type stars is expected to be around 5-10 AU for the core-accretion process. HR 8799 is somewhat more massive and luminous than the Sun which will affect this somewhat, not necessarily in a simple manner either as volumes of material, orbital velocities and the speed at which the star evolves through the pre-main sequence stage will all play a part. The other proposed mechanism for forming gas giants is gravitational instability, which (if it indeed is a valid formation mechanism) ought to work at much larger distances, as the gas located too close to the star would not cool down fast enough for instability to occur. HR 8799e is therefore problematic because it is too far out for core accretion, and too close to the star for gravitational instability. There is also an asteroid belt located just inside its orbit, which probably puts strong constraints on planet migration models as well. Even without this, explaining how the HR 8799 system managed to form so many very massive planets is challenging enough already!

  25. Messier Tidy Upper

    @ ^ Andy : Great comment. Thanks. :-)

    @25. Isaac : The orbit of Uranus is close to the Sun?

    Relatively. In this case. Compared to Pluto or Sedna or the cometary Oort cloud – yes. Compared to Jupiter or Mercury, natch, not-so-much. ;-)

    Just how far out are large planets expected to form?

    Expected? I’m not sure we know *what* to expect! Exoplanets keep surprising us and making us think again about what is & isn’t possible.

    Observationally, we’ve found :

    * 2M 1207b, the first exoplanet photographed (2004 April 27th) orbits 55 AU from its brown dwarf sun. Some I think have argued that it may itself just possibly be a smaller brown dwarf in part for that very reason but still.

    * GQ Lupi b & AB Pictoris b are rival claimants for that “first exoplanet imaged” honour but both are probably brown dwarfs rather than superjovian gas giants. Both have uncertain high~ish masses orbiting hundreds of AU from a protostar and a K2 orange dwarf respectively.

    * Fomalhaut b orbits at 4 times Neptune’s distance (30 AU x 4 or 120 AU if my dodgy maths is right) from its A type Sirian sun.

    So exoplanets can certainly be found at distances *much* further from their primary stars than we find in our Solar system.

    Oh also :

    HAPPY NEW YEAR! :-D

    Hope y’all have a excellent 2011. :-)

    (Heading towards early evening on New Years Day in my time zone, btw.)

NEW ON DISCOVER
OPEN
CITIZEN SCIENCE
ADVERTISEMENT

Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!

ADVERTISEMENT

See More

ADVERTISEMENT
Collapse bottom bar
+

Login to your Account

X
E-mail address:
Password:
Remember me
Forgot your password?
No problem. Click here to have it e-mailed to you.

Not Registered Yet?

Register now for FREE. Registration only takes a few minutes to complete. Register now »