Weather sizzles on a planet that kisses its star

By Phil Plait | January 28, 2009 11:00 am

Still suffering from the heat, the planet reaches the peak of its orbit, momentum carrying the bizarre world far, far from its star.

From that distance the star, slightly less massive and cooler than our own Sun, would gently heat the planet, giving it an almost Earthlike temperature. And during the days and weeks it spends out there, the planet cools. But it’s not enough. It’s never enough.

HD 80606b, the Icarus planet.

The planet has pulled away as far from the star as it can, but gravity cannot be denied. Slowly, inexorably, it begins the long fall back. Growing ever-larger as the planet draws ever-nearer, the star swells hugely. Just 55 days after reaching its farthest distance from the star, the planet drops hellishly close to the star’s surface, and a furnace blast of heat bears down on it. The planet’s atmosphere roils and churns, reaching temperatures hot enough to melt copper.

Accelerated by the star’s gravity to a speed hundreds of times faster than a rifle bullet, the planet whips around the star and begins the long climb back. It will cool down as it puts over 100 million kilometers between itself and the fierce light of the star… but the cycle will repeat itself, and the planet will burn once again.


orbit diagram of HD 80606b
The orbit of HD 806060b. The small
circles represent 1 hour intervals,
the four positions separated
by 19 hours each.

Such is the life of HD 80606b, a gas giant planet four times the mass of Jupiter that orbits a star 190 light years from Earth. The planet’s orbit is incredibly elliptical, with a whopping eccentricity value of 0.927 — meaning the orbit is elongated like a rubber band being fought over by jealous children. It’s posited that gravitational interaction over time with a distant binary stellar companion to the star may have forced the orbit into this shape; it peaks at a distance of 125 million kilometers (75 million miles) from the star, but the planet’s 111 day orbit drops it to a mere 4 million kilometers (2.4 million miles) from the star’s surface. In the 55 days it takes to drop, it sees the disc of the star swell to 30 times its previous size, flooding the planet with nearly 800 times the amount of heat it felt at greatest distance.

During the close approach in late November, 2007, astronomers used the Spitzer Space Telescope to observe the planet. They couldn’t resolve the planet; in fact they couldn’t even see it at all. All they could detect was the slight increase in infrared light emitted by the warming planet adding to the star’s light; a tiny fraction to be sure, but detectable by Spitzer. In just six hours, the planet’s upper atmosphere heated from about 500 C to over 1200 C (980 to 2,240 degrees Fahrenheit).

Using very sophisticated computer models, the scientists could make an image representing what the planet’s upper atmosphere would look like after its dip into hell. The image above is a simulation, a computer model of this Icarus planet 4.4 days after closest encounter. If you had infrared eyes and hovered over the planet, it might look something like this.

The blue light is the day side of the planet, lit by reflected light of the star. Most of the planet we see here is facing away from the star, so it’s night. But it’s hardly dark: the red glow is actually the heat from a massive storm, blasted into life by the intense energy absorbed during the close passage. It’s almost an explosion, with supersonic winds whipping air at 5 kilometers (3 miles) per second from the day side around to the night side. The planet’s rotation causes a Coriolis effect which wraps the storm into curls as it screams away from the heat of the star.

Think on it: this map shows violent weather on a distant planet we’ve never actually seen.

While this isn’t a real image of the planet, it represents a huge step in our understanding of these hot gas giants which orbit so close to their stars. Remember, that’s not an artist’s drawing; it’s the output of a computer model of the atmospherics of the planet. The observations using Spitzer give astronomers critical information on how much the planet is heated, and how. Armed with that information, and the ability to understand how atmospheres behave, they can generate models like these.

While no other planet we know of makes such an extreme path around its star, there are plenty of other planets out there to observe. And we may find another bizarre world like this one; in fact, it’s almost certain we’ll find many. We’ve only detected the presence of about 300 planets, and there must be tens or even hundreds of billions of planets orbiting stars in the Milky Way.

I wonder: what’s out there that will make this planet seem almost normal by comparison?

Planet image credit: D. Kasen, J. Langton, and G. Laughlin (UCSC). Orbital diagram from their paper in Astrophysical Journal.

CATEGORIZED UNDER: Astronomy, Cool stuff, Science

Comments (79)

  1. “I wonder: what’s out there that will make this planet seem almost normal by comparison?”

    Perhaps a planet where apes evolved from men?

  2. Rob

    Very cool.

    Phil, I have very much enjoyed your interactions with The Skeptics Guide to the Universe rogues… please keep it up!

    Rob

  3. IVAN3MAN

    It appears that the I.A.U. is gonna have to redefine what constitutes a “Planet”.

  4. knobody

    when i was my son’s age, other planets were still hypothetical. almost certain, but unknown. reality is so far more fascinating than mythology that i wonder why people still look to the stars and talk about pretend heavens.

  5. CR

    At first glance, I thought the picture was from Space: 1999… it resembles many of the planets that show used.

    On a more scientific track, at what point will this planet have its atmosphere blasted away by such repeated close passes to its star? Or is the planet’s massive size enough to allow it to retain its atmosphere? I’m guessing that the lees energetic output of the star may help, as well. But STILL, the thing gets closer to its star than Mercury does to our sun! Yikes!

    Once again, the real universe is far stranger (and more wonderful) than fiction…

  6. CR

    lees = less

    Preview function! (I know, I know, proofread…)

  7. Jim Roberts

    Hal Clement should still be alive to write a novel about life on that planet!

  8. John

    “I wonder: what’s out there that will make this planet seem almost normal by comparison?”

    I’m not a scientist, so I might be wrong, but given what we know about all the planets in our solar system, and all the extrasolar planets we’ve discovered, wouldn’t Earth fit this bill?

  9. Bandsaw

    That’s a cometary eccentricity on a planet larger than any in our solar system. I wonder if there is anything else left orbiting that star, or if this monster ate or ejected everything else.

  10. exoplanets.org is listing the following planets with an eccentricity greater than 0.80. That would mean stellar irradiance at periastrion is at least 81 times stellar irradiance at apastrion.

    HD4113 b eccentricity = 0.903
    HD20782 b eccentricity = 0.925
    HD80606 b eccentricity = 0.935
    HD156846 b eccentricity = 0.847

    So that’s 4 out of 228 exoplanets or about 1.75 percent. If you relax the criterion to eccentricity greater than 0.70 (periastrion irradiance greater than 32 apastrion irradiance) you pull in a few more:

    HD2039 b eccentricity = 0.715
    HD37605 b eccentricity = 0.737
    HD41004A b eccentricity = 0.74
    HD45350 b eccentricity = 0.798
    iota Dra b eccentricity = 0.712
    HD187085 b eccentricity = 0.75
    HD222582 b eccentricity = 0.725

    So that’s 11 exoplanets out of 228 (~5 percent) with an eccentricity above 0.7. So extremely eccentric orbits are not that unusual among the exoplanets discovered to date.

    Also, the dwarf planet Sedna in our own solar system has an eccentricity of 0.855 with a perihelion distance of 76 AUs but an aphelion distance of a whopping 975 AUs (maximum solar irradiance equals 164 times minimum solar irradiance).

  11. And just to hit home how LITTLE of the galaxy we have really found planets to look at, check out this image: http://forums.randi.org/imagehosting/9574973c8a6aebf9.jpg (direct link in case the HTML doesn’t stack up)

    We have so much more to explore!

  12. I agree with John, the weirdest planet I can think of is the one I’m sitting on. Everyone else is normal by comparison.

  13. Jason Heldenbrand

    I don’t know about you guys, but this stuff gives me -loads- to work with on my Star Wars tabletopping group. Player characters have to escape planet before it becomes a raging, boiling inferno! Keep it coming astronomers, this GM is tired of making this stuff up.

  14. JackStraws

    “Accelerated by the star’s gravity to a speed hundreds of times faster than a rifle bullet…” For comparison’s sake, after doing some rough calculations, I found that the Earth’s orbital speed is somewhere in the neighborhood of only twenty times the speed of a reasonably fast rifle bullet.

    So yeah, I’d hate to get in the way of this thing, that’s for sure.

  15. IVAN3MAN

    Dr. Phil Plait:

    In the 55 days it takes to drop, it sees the disc of the star swell to 30 times its previous size, flooding the planet with nearly 800 times the amount of heat it felt at greatest distance.

    Excuse me for nit-picking, Phil, but according to inverse-square law where I is the intensity of radiation at distance r:

    I1 = I2 × r22 × [1 ⁄ r12]

    Let: I2 = 1 and r2 = 125M km at aphelion; r1 = 4M km at perihelion.

    Therefore, I1 = 976.5625

  16. IVAN3MAN

    Furthermore, if one had a bloody preview/edit facility here, I would have spaced that out better!

  17. American Voyager

    Sometimes science fiction does a good job of guessing. I remember watching an episode of Lost in Space as a boy 40 years ago where they were on a planet with an orbit like that. Yes, it was terrestrial, but they had a scene where the sun “passed across the sky” and much of the plantlife burst into flames.

  18. @IVAN3MAN “Therefore, I1 = 976.5625″

    A better equation for maximum irradiance divided by minimum irradiance would be:

    Imax/Imin = ((1 + ecc) / (1 – ecc))^2

    The original eccentricity referenced by Phil for HD80606 b was 0.927. Thus, Imax/Imin = 696.8

    Exoplanets.org is listing an eccentricity of 0.935 so Imax/Imin = 886.2. So Phil’s estimate of Imax/Imin = 800 is matching the Exoplanets.org data better.

  19. You have to wonder — if exoplanets can be in such eccentric orbits, are there any that are or were rogue interstellar planets? Statistically, it seems likely that a few might exist. After all, if comets can have parabolic trajectories, why not planets too?

  20. IVAN3MAN

    @ Tom Marking,

    Well, I was just going by the distance figures for aphelion/perihelion of HD 80606b that Dr. Plait stated in his post. :|

  21. WJM

    You know those planets that are, like, double planets?

  22. IVAN3MAN

    Larian LeQuella, I checked the source code, and the reason why you did a SNAFU on your attempt to post that image was because you had neglected to enclose the URL within double-quotation marks. Silly boy! :P

  23. LOL!!! Man, I suck at typing… Well, at least I know my limitations and provided the direct link as well. SItuation Normal!

    Again, another reason for an edit feature! ;)

  24. IVAN3MAN

    Larian LeQuella, maybe you can ask Phil Plait to fix it for you.

  25. I remember seeing on a past exam paper a question along the lines of “give 3 differences between the orbit of a planet and the orbit of a comet”. One of the answers the mark scheme wanted was that planets go in circular orbits, and comets are in highly elliptical ones.

    All hail the Kozai mechanism.

  26. Cheyenne

    Larian – Wow, that pic puts things in perspective. That is cool. I actually thought we were finding planets throughout the Milky Way. Looks like there is a lot of real estate out there to explore (to put it mildly).

  27. CR Says: “At first glance, I thought the picture was from Space: 1999… it resembles many of the planets that show used.”

    Right era, wrong show. It’s not “1999” but “Lost in Space.” If you remember (at least those of a certain age) the first few episodes after landing on the planet the first season, the orbit carried it far from the “sun” (that’s what they called it) and then in on a swooping pass that got things hot enough to ignite the brush. Supposedly, the planet’s “year” was about two Earth-weeks long, based on the time in the episodes from deep freeze to incineration. Strangely, this odd weather was never mentioned again the rest of the season.

    – Jack

  28. T.E.L.

    Jack,

    CR wasn’t talking about the planet’s eccentric orbit. It was the planet’s quasi-psychedelic color scheme, which fits the bill for Space: 1999.

  29. gopher65

    If they don’t name this thing Icarus, they need their heads examined;).

  30. CR

    Interesting fact about Lost in Space, Jack, but T.E.L. is correct in what I meant. (On the other hand, it’s cool that an earlier show came up with a planet that had such an eccentric orbit. Too bad they didn’t stay consistent with continuity, but at least the original idea was neat!)

  31. Thanks BA. Great post, awesome exoplanet! :-D 8) :-)

    @ andy who says on Jan 28th, 2009 at 2:49 pm :

    … One of the answers the mark scheme wanted was that planets go in circular orbits, and comets are in highly elliptical ones. All hail the Kozai mechanism.

    Er .. what’s that? Hadn’t heard of that ‘un.

    @ Larian LeQuella saying :

    LOL!!! Man, I suck at typing ..

    If * you * suck at typing then I hate to think where that puts me! ;-)

    Seriously BA when is this editing / previewing capability getting here – it just can’t come soon enough for my liking.

    From Adelaide, South Australia where it is 42 degrees celcius (well over 100 Fahrenheit) outside at the moment .. :-(

    Still at least this exoplanet puts that sweltering heat in perspective!

    Doesn’t make it feel any cooler but .. ;-)

    PS. I agree 100% with with gopher65 who said :
    (January 28th, 2009 at 5:15 pm)

    “If they don’t name this thing Icarus, they need their heads examined.” ;-)

    Absolutely – when is the IAU looking at official names for significant exoplanets? Anyone know? Its about time if y’ask me! ;-)

    —–

    Without our atmosphere the Earth’s average temperature would be minus eighteen degrees Celsius.
    – Dr Alan Longstaff, ‘Astronomy Now’ magazine July 2007.

    Procyon, the “Little Dogstar” and 8th brightest star in our skies, emits as much light in a day as our Sun does in a week.
    – Ken Croswell, article on Procyon, (Alpha Canis Minoris) (see link to my name above.)

    Today is one of those days when I wish our Sun was less like Alpha
    Centauri A* and more like Alpha Centauri B! +

    * Alpha Centauri A = a G2 yellow dwarf, 1.7 x the Sun’s brightness,
    just a bit larger (10 % more massive), hotter and brighter than our
    daytime star.

    + Alpha Centauri B = a K2 orange dwarf, a bit dimmer, cooler and
    smaller than our Sun. It’s mass is 85 % of our Suns and it has 40 %
    of the Sun’s brightness.
    – Source : Page 15, Kaler, James B., “The Hundred Greatest Stars”,
    Copernicus Books, 2002.

    (Not Alpha Cen C or Proxima Centauri though – as an M5 red dwarf that’s a bit too cold for my liking even now when dying of frostbite seems like bliss! ;-)

  32. Joker

    If that’s a planet kissing it’s star then I hate to think how hot a planet ******* it’s star would be! ;-)

  33. Joker

    Sorry, that was a bit crude even for me! ;-) :-(

  34. Jim Roberts: He did, “Cycle of Fire”.

  35. Jim Roberts

    # Edmund Schluessel Says:

    Jim Roberts: He did, “Cycle of Fire”.

    Sort of, but the cycle there is much longer than the orbital period of this planet, so that at least the cold people get quite a long lifetime between hot phases. On this planet, the natives would have to manage to survive through the whole temperature range, though I suppose they could cheat by burrowing deep into the crust and estivating.

  36. Ebenezer

    Wow, that’s nearly as hot as Melbourne!

  37. Moose

    Pern may well be in _that_ system.

  38. MarkW

    “If they don’t name this thing Icarus, they need their heads examined.”

    There’s already a body called Icarus in our Solar System: http://en.wikipedia.org/wiki/1566_Icarus

  39. Jeff

    So this planet is able to sling-shot itself back in time, and save its own whales… cool.

  40. Gary Ansorge

    So, the difference between a planet and a comet is that the planet is a whole lot bigger???
    Shouldn’t this planet have a tail?

    Earths greatest peculiarity(aside from having carbon based life forms) is that it is a binary system. I wonder,,, how long before we find another such example?

    Remind me again,,,where is the common center of rotation of the Earth Luna system???

    Gary 7

  41. SD

    Dude, watch the comma overuse there.

  42. @Gary “Remind me again,,,where is the common center of rotation of the Earth Luna system???”

    About 4,700 km from the center of the earth which is still 1,600 km (1,000 miles) beneath the earth’s surface.

  43. T.E.L.

    What this once again points out is that the whole notion of “planetness” is a paradigm that died once the telescope was invented. People keep trying to freshen up the corpse, but it’s been rotting all along and is starting to stink up the joint.

    We need to grow up and abandon the planets. There are no planets. There are no moons. There are gravitating bodies of varied chemical, geological, etc compositions, with varied orbital elements with respect to each other. The time is getting ripe for a Hertzsprung-Russell type organizational scheme (or, alternatively, a “periodic table of the elements” type scheme) for the variety of objects which populate the Universe. A meta-scheme would even include the HR diagram for stellar bodies as a special subset.

  44. Chris A.

    @Cheyenne:

    “Larian – Wow, that pic puts things in perspective…Looks like there is a lot of real estate out there to explore (to put it mildly).”

    Keep in mind that the picture Larian posted is illustrative only–not even close to scale. The diameter of the sphere encompassing most of the known exoplanets is less than 1/300th the diameter of the galaxy.

  45. Chris A.

    Whoops, make that 1/170th. (I did radius instead of diameter. Mea culpa.)

  46. JoBrad

    So, like CR, I’m wondering: Why does this planet have an atmosphere? I thought one of the reasons that the planets closer than Earth to the Sun in the Solar System was because of the solar wind?

  47. Er .. what’s that? Hadn’t heard of that ‘un.

    Basically for objects in orbits with high mutual inclination, gravitational interactions cause the objects to exchange eccentricity and inclination – this is termed the Kozai mechanism. HD 80606b may have been born into a low eccentricity orbit which had the misfortune to be highly inclined with respect to the binary orbit of the HD 80606+HD 80607 system. (There’s no particular guarantee that protoplanetary discs in a wide binary system will be aligned to the system’s orbit.) The Kozai mechanism would have then caused the eccentricity to increase until the planet got so close to the star at periastron that tidal forces disrupted the interactions, effectively parking the planet in its current extremely elliptical orbit. It is interesting to note that the planets with the highest eccentricities are all in wide binary systems. This suggests that the extreme orbits are a result of this process.

  48. Torbjörn Larsson, OM

    I wonder,,, how long before we find another such example?

    What about Pluto-Charon?

  49. @andy “Basically for objects in orbits with high mutual inclination, gravitational interactions cause the objects to exchange eccentricity and inclination – this is termed the Kozai mechanism. HD 80606b may have been born into a low eccentricity orbit which had the misfortune to be highly inclined with respect to the binary orbit of the HD 80606+HD 80607 system.”

    Wow, I actually learned something new. So how does this Kozai mechanism actually work? Apparently HD 80607 is located only 1,200 AUs from HD 80606. The two of them are sometimes called the Struve 1341 binary star system.

    Considering the 3 other exoplanets with eccentricities above 0.8:

    HD4113 b eccentricity = 0.903
    No definite stellar companion has been found. There is speculation that it might have either a brown dwarf or faint white dwarf as a companion.

    HD20782 b eccentricity = 0.925
    I can’t seem to find any information about a binary star for this one.

    HD156846 b eccentricity = 0.847
    HD156846 seems to have a red dwarf companion called IDS 17147-1914B. I couldn’t find any distance information on it.

    Perhaps the Kozai mechanism is not the only thing that can cause high eccentricities in planetary orbits.

  50. StevoR

    @ andy Thanks for the explanation – first I’ve heard of it! :-)

    Gary Ansorge asked :
    (January 29th, 2009 at 7:44 am )

    “So, the difference between a planet and a comet is that the planet is a whole lot bigger??? Shouldn’t this planet have a tail?”

    Among a few other things yeah, size is a big difference there! ;-)

    @ Tom Marking : Interactions with other massive exoplanets maybe?

    @ MarkW D’oh! That’s true – I forgot all about asteroid Icarus. Maybe we could start having a few names applied to more than one object -context should make it easy to tell them apart. Or not.

    Anyway “catalogue” names are a personal gripe of mine – I hate them and think that at least significant exoplanets and their stars should be awarded common names for ease of memorisation and communication.

    Ebenezer says :

    (January 29th, 2009 at 5:08 am )
    “Wow, that’s nearly as hot as Melbourne!”

    Yes, but we’ve alreday had three days of 40 plus incl. the third hottest day on record and hottest night ever here in Adelaide. Its 40 + now. ;-(

  51. StevoR-Testing

    Andy also wrote :

    “It is interesting to note that the planets with the highest eccentricities are all in wide binary systems. This suggests that the extreme orbits are a result of this process.”

    So would this be a big problem for any hypothetical planets around Alpha Centauri – or, for that matter, the V-lizards from a planet of Sirius mentioned on another thread?

    @ T.E.L. :
    (January 29th, 2009 at 11:27 am)
    What this once again points out is that the whole notion of “planetness” is a paradigm that died once the telescope was invented. People keep trying to freshen up the corpse, but it’s been rotting all along and is starting to stink up the joint.

    We need to grow up and abandon the planets. There are no planets. There are no moons. There are gravitating bodies of varied chemical, geological, etc compositions, with varied orbital elements with respect to each other. The time is getting ripe for a Hertzsprung-Russell type organizational scheme (or, alternatively, a “periodic table of the elements” type scheme) for the variety of objects which populate the Universe. A meta-scheme would even include the HR diagram for stellar bodies as a special subset.

    Well I don’t think we’ll *ever* get rid of the term or idea of planets.

    I do think we should categorise diferent types of planets and have an inclusive definition for thegenreal category of “planet” which is thenbrojken down into various types or kinds eg. rocky Earth-like terrestrial planets, gas gianst, ice dwrafs, Hot Jupiter’s, Supe-Neptunes, Super-Venus planets, etc .. like we do for stars eg. Blue dwarfs (eg. Regulus), blue giants (eg. Achernar), blue supergiants (eg. Rigel), white supergiants (eg.Deneb), yellow dwarfs, yellow supergiants (eg. Alpha Aquarii, Canopus, Delta Cephei), red supergiants, red dwrafs, White dwrafs, Sirian (A-type) stars, Procyonese (F-Type) dwarfs, et cetera.

    I think that’s the way the IAU should have gone at Prague 2007 with the whole Pluto-Eris planets controversy and hope they will review and change their decision to take that sort of path.

  52. StevoR-Correcting

    Make that :

    Well I don’t think we’ll *ever* get rid of the term or idea of planets.

    I do think we should categorise diferent types of planets and have an inclusive definition for the general category of “planet” which is then broken down into various types or kinds of planet :

    Eg. rocky Earth-like terrestrial planets, gas gianst, ice dwrafs, Hot Jupiter’s, Supe-Neptunes, Super-Venus planets, etc ..

    In the same way we do for stars eg. its a star then its broken into a particular type or class :

    Eg. pulsar, neutron star, magnetar, Blue dwarf (eg. Regulus), blue giants (eg. Achernar), blue supergiants (eg. Rigel), white supergiants (eg.Deneb), yellow dwarfs, yellow supergiants (eg. Alpha Aquarii, Canopus, Delta Cephei), red supergiants, red dwrafs, White dwrafs, Sirian (A-type) stars, Procyonese (F-Type) dwarfs, et cetera.

    I think that’s the way the IAU should have gone at Prague 2007 with the whole Pluto-Eris planets controversy and hope they will review and change their decision to take that sort of path.

  53. StevoR-Correcting

    Yeah iknow white dwrafs and pulsars, magnetars, neutron stars etc .. areactually dead stars thatdon’t meet theshine by nuclearfusion criteria.

    Brown dwarfs – often called “failed stars” could equally well be termed “really successful Jupiters” and incl. under the planets category ..

    Sure there are a lot of complications and issues but the basic idea of planets – no, I can’t see that disappearing.

  54. StevoR-Correcting

    Typos! Sigh.

    I know I should ‘ve used word and then spell-checked and thenposted but that’s a nuisance too.

    I’d really love an editing /previewing and even deleteing posts ability here Bad Astronomer. I’ve heard one is coming but it never seems to get here. Really, it shouldn’t be that hard to do and other blogs have had such facilities for years .. Please, Dr Phil Plait, please. It’d just make this blog so much better.

  55. IVAN3MAN

    @ StevoR,

    Yes, indeed, like they have on RichardDawkins.net.

  56. Tom Marking: regarding the binarity of HD 20782, check out arXiv:astro-ph/0610623 for details of the stellar companion, however the estimated timescale for Kozai oscillations induced by the observed companion appears to be too long (several Hubble times). Nevertheless it does fit the pattern of binary systems. As for HD 4113, there is evidence for an outer companion from the radial velocity residuals to the 1-planet solution (see arXiv:0710.5028), however the companion has not been characterised.

  57. Assuming there is a large moon orbiting around HD80606 b with an albedo and greenhouse temperature increase similar to the earth, an inhabitant of this moon might issue the following report:

    Day 1: It’s apastrion. We are 0.9056 A.U. from our star travelling at 8.49 km/sec. Stellar flux is only 1,166 watts per square meter. The temperature outside is a chilly 5.7 degrees Celsius.

    Day 7: We are now 0.8987 A.U. from our star travelling at 9.59 km/sec. The temperature is still chilly, only 6.6 degrees Celsius.

    Day 24: We’re getting a little bit closer, only 0.8008 A.U. from our star travelling 19.41 km/sec. Stellar flux is now 1,491 watts per square meter. Average surface temperature on our moon is now 20.8 degrees Celsius. We can go outside without a jacket. It’s wonderful weather.

    Day 30: We’re 0.7347 A.U. from our star travelling at 24.71 km/sec. Stellar flux is now 1,771 watts per square meter and the temperature has risen to 32.0 degrees Celsius. It’s beginning to get a little warm now.

    Day 34: We’re 0.6795 A.U. from our star travelling at 29.02 km/sec. Stellar flux is now 2,071 watts per square meter and the temperature is now 42.6 degrees Celsius. It’s now uncomfortably hot. We have to run our air conditioners at full speed just to keep cool.

    Day 37: We’re 0.6311 A.U. from our star travelling at 32.86 km/sec. Stellar flux is now 2,400 watts per square meter. The temperature has climbed to 52.9 degrees Celsius. Our air conditioners have given out and we have had to retreat into the recesses of deep caves just to stay warm.

    Day 45: We’re 0.4647 A.U. from our star travelling at 47.91 km/sec. Stellar flux is now 4,428 watts per square meter. The temperature has climbed to 100.3 degrees Celsius and our oceans have started to boil away. It is too hot now even in the caves and we have decended deep underground just to survive the heat.

    Day 48: We’re 0.3514 A.U. from our star travelling at 62.03 km/sec. Stellar flux is now 7,745 watts per square meter. The average surface temperature of our moon is now 150.3 degrees Celsius. The oceans have long since boiled away. We have to keep burrowing deeper and deeper into the crust to avoid the intense heat at the surface.

    Day 54: The day before periastrion. We’re only 0.0902 A.U. from our star travelling at 121.1 km/sec. Stellar flux is now 117,500 watts per square meter and average surface temperature is 523.5 degrees Celsius which is hot enough to melt lead and zinc. We are now several hundred meters beneath the crust just to survive.

    Day 55: The day of periastrion. We come to within 0.0304 A.U. of our star travelling at 182.46 km/sec. Stellar flux reaches a maximum of 1,033,000 watts per square meter. The average surface temperature is now 1,070.0 degrees Celsius which is hot enough to melt silver and gold. Looking forward to the cooler weather ahead.

    Such is the life of the HD80606 b-ers.

  58. Ooops, there’s a slight correction in the orbital velocities as follows. Everything else is correct:

    Day 1: 8.48 km/sec
    Day 7: 9.43 km/sec
    Day 24: 19.02 km/sec
    Day 30: 24.22 km/sec
    Day 34: 28.44 km/sec
    Day 37: 32.17 km/sec
    Day 45: 46.61 km/sec
    Day 48: 59.70 km/sec
    Day 54: 141.72 km/sec
    Day 55: 252.54 km/sec

  59. Assuming there is a large moon orbiting around HD80606 b

    Not a good assumption, the dynamics of satellite orbits around the planet are extremely unstable.

    According to Domingos, Winter and Yokoyama, 2006, MNRAS 373, 1227, the formula for the outermost stable prograde orbit as expressed as a fraction of the Hill radius is 0.4895 * (1 – 1.0305*e_p – 0.2738*e_s) where e_p and e_s are the planet and satellite eccentricities. The Hill radius fraction for retrograde orbits is 0.9309 * (1 – 1.0704*e_p – 0.9812*e_s). The Hill sphere itself is given by a*(1-e_p)*(m/(3*M))^(1/3) where a is the planet’s semimajor axis, m is the planetary mass and M is the stellar mass.

    Taking e_s is zero (for maximum size of the stable zone), for HD 80606b this fraction works out as 0.0219, which when converted to physical units by multiplying by the size of the Hill sphere, you get a distance of 11,000 km, which is rather smaller than the typical radius of a jovian planet (~70,000 km or so). The retrograde case equates to an even smaller fraction for the eccentric orbit of HD 80606b – note that for circular orbits, the retrograde satellite region is larger. There is likely nothing in orbit around the planet in the way of moons or ring systems.

  60. @andy “There is likely nothing in orbit around the planet in the way of moons or ring systems.”

    Yeah, too bad. The moon was obviously a literary device to imagine what would happen to an Earth-like planet in such an orbit.

  61. Yeah. On the other hand, it’s not every planet that has a fiery supersonic shockwave blasting through its atmosphere every orbit… might look pretty impressive (though obviously you wouldn’t hear it coming…)

  62. StevoR

    Thanks for that journal of life around the most eccentric exoplanet Tom Marking & thanks again andy for the info that it won’t have any such moon or rings and why. Seems like a bit of a waste doesn’t it!

    SF Author Brian Aldiss wrote a series of books on a hypothetical planet called ‘Helliconia’ that must have been in a similar though somewhat less extreme orbit. Helliconia was in a binary star system with a hot white star and red dwraf (from memory) and had a very long orbit taking it near first one then the other sun with “seasons” lasting hundreds of years – of extreme cold then extreme heat. The (3?) Helliconia series books were each set in a season separated by hundreds if not thousands of years – Helliconia Winter, Spring, Summer & if I recall right. Good reads & had a lot of very interesting and wellthought out ‘world creation’ astronomy and mythology, plus a science-religion clash theme in ‘em too ..

  63. StevoR

    Can’t we all just call it the Struve 1341 binary star systemthen given that’stheshortest and most easilymemorised name?

    Indeed, I suggest astronomers adopt a rule with catalogue names that where one star seems to have many different catalogue names that the one used is always the shortest and thus easiest – and that proper names be given to significant stars too! Proper names should always be used in preference to numerical designations if y’ask me! It makes it so much easier for laypeople to remember and to discuss with them.

    After all, how do you pronounce HD 80606 b? Single digits eg. “eight-zero-six ” etc ..? The entire number? “Eighty thousdnad six hundred” etc .. Some split up variant along the lines of “eighty sixty sixty”?

  64. StevoR

    Question : Would this exoplanets extremely eccentric orbit rule out any other planets in the system?

    Would it demolish the possibility of other asteroids, comets etc .. around its star?

    Can we say this presumably lifeless ball of superheated hydrogen is pretty much certainly the only object in the system aside from its star?

    andy says on January 30th, 2009 at 3:31 pm :

    Tom Marking: regarding the binarity of HD 20782, check out arXiv:astro-ph/0610623 for details of the stellar companion, however the estimated timescale for Kozai oscillations induced by the observed companion appears to be too long (several Hubble times). Nevertheless it does fit the pattern of binary systems. As for HD 4113, there is evidence for an outer companion from the radial velocity residuals to the 1-planet solution (see arXiv:0710.5028), however the companion has not been characterised.

    Any idea if this uncharacterised outer companion for HD 4113 is planetary, “brown dwarfery” or stellar in nature?

  65. Also, using the more realistic value of 111.45 days for the orbital period of HD 80606 b one can compute the season lengths. Assuming some sort of axial tilt for the planet, then if the summer solstice point (northern hemisphere) is the same as the periastrion point we come up with the following season lengths:

    Duration of summer = 0.55 days
    Duration of autumn = 55.17 days
    Duration of winter = 55.17 days
    Duration of spring = 0.55 days

    Spring and summer together last barely longer than 1 day whereas autumn and spring together last more than 110 days.

  66. Concerning the Kozai mechanism and HD80606 b, from Wikipedia it says that:

    cos(i) * sqrt(1 – ecc^2) = k

    i is inclination angle
    ecc is eccentricity
    k is a constant which is preserved throughout time

    Assuming HD80606 b is now in the same plane as the orbit of HD80607 (not sure if this is the case) then i = 0:

    cos(0) * sqrt(1 – 0.9349^2) = 0.3549 = k

    At some previous time we would have:

    i = arccos(k / sqrt(1 – ecc^2))

    Let’s assume that previously ecc was 0.0 (perfectly circular orbit). Then

    i = 69.2 degrees

    HD 80606 b could have started out with a perfectly circular orbit inclined 69 degrees relative to HD 80607. The article doesn’t say how long it takes for the Kozai mechanism to occur.

    I’m not sure what the Kozai angle of 39.2 degrees is all about.

  67. @andy “Tom Marking: regarding the binarity of HD 20782, check out arXiv:astro-ph/0610623 for details of the stellar companion”

    Yes, lots of good stuff here.

    http://www.citebase.org/fulltext?format=application%2Fpdf&identifier=oai%3AarXiv.org%3Aastro-ph%2F0610623

    “HD 20782 is listed as a binary in the CCDM catalog (CCDM 03201-2850). The companion is HD 20781. The very large separation, 252 arcsec, that corresponds to 9080 AU at the distance of HD 20782 (36 pc) makes a detailed check of physical association mandatory. … Therefore, HD 20782 and HD 20781 form a very wide common proper motion pair. Considering the nominal radial velocity and proper motion differences, the pair could be bound…

    Only two extrasolar planets have orbits with eccentricities larger than 0.8: HD 20782b and HD 80606b. Both their host stars are member of wide common proper motion pairs. While the link between the high eccentricity and binarity still needs statistical confirmation, nevertheless it is interesting to investigate the possible ways in which a distant companion might have induced extreme planet eccentricities.

    As discussed in Sect. 4, the extremely high eccentricity of HD20782b is unlikely to be due to the Kozai mechanism, as the period of the eccentricity oscillation is much larger than the Hubble time.

    An interesting possibility to explain both the very large separation of the binary and the very high eccentricity of the planet orbiting the primary is a dynamical encounter of the binary (or of an originally higher multiplicity system) within a star cluster or with a passing star, that might have perturbed both the binary and the planet orbit. According to the simulations byWeinberg et al. (1987), the probability of survival for a binary with an initial semimajor axis of about 10000 AU after 4 Gyr is less than 50%.”

    http://www.astro.northwestern.edu/rasio/Papers/139.pdf

    also has some good stuff on this:

    “Of the 86 planets with orbital periods greater than 10
    days and best-fit orbital eccentricities exceeding 0.2, two
    planets have eccentricities that are currently estimated to
    be greater than 0.8, HD 80606b (e = 0.935±0.0023; Naef
    et al. 2001) and HD 20782b (e = 0.925±0.03; Jones et al.
    2006). Such large eccentricities are unlikely to be the re-
    sult of planet–planet scattering (at least in the context of
    two planets initially on nearly circular orbits, as explored
    in §3.2.2). Thus, we search for alternative explanations
    for these two planets with extremely large eccentricities…

    The current estimates in both of these systems are quite large (Desidera & Barbieri 2006). This has led to speculation that the “Kozai effect” may not be able to explain the large eccentricities for these two systems. The bina-
    rity may still be significant, e.g., if the two stars were not born as a binary, but rather the current binary companion originally orbited another star and was inserted into a wide orbit around the planetary system via an exchange interaction (a formation scenario similar to that proposed for the triple system PSR 1620–26; Ford et al. 2001; or other putative planets in multiple star systems; Portegies Zwart & McMillan 2005; Pfahl & Muterspaugh
    2006; Malmberg et al. 2007). During such an encounter, the four-body interactions might have induced a large eccentricity in the planetary orbit. Such interactions may have been common for stars born in clusters or other dense star forming regions (Adams & Laughlin 1998; Zakamska & Tremaine 2004).”

    So maybe the Kozai mechanism isn’t the reason afterall.

  68. Tom Marking:
    A better summary than Wikipedia’s, with relevant formulae is at orbitsimulator.com/gravity/articles/kozai.html, which includes timescales and maximum amplitude of the oscillations. The Kozai angle is the angle which the mutual inclination has to exceed for Kozai oscillations to occur. If I’m doing the maths right, assuming the inclination of the planet with respect to the binary is randomly distributed (which may be the case in binaries where the separations are of the order hundreds of AU or more), the angle will exceed the Kozai angle approximately 77% of the time. This also has implications for the survival of planetary systems if another star is exchanged into the system: even if the new star is far away enough that things would otherwise be fine, the majority of the time the system will end up having a mutual inclination exceeding the Kozai angle.

    Regarding the conclusion of that paper you linked about Kozai being ruled out for HD 80606, the paper cited for that conclusion states that Kozai oscillations would be able to explain HD 80606b’s orbit for inclinations greater than 85 degrees or so, which corresponds to a probability of about 9%, so not an extremely improbable scenario. However HD 20782b (which recent estimates give an eccentricity of 0.97 !!!) cannot be explained by Kozai oscillations due to the known companion HD 20781 because of the timescale. What would be really useful for solving the mystery of these extremely eccentric planets are measurements of the Rossiter-McLaughlin effect, which requires a transit. Unfortunately this is not particularly likely for HD 80606b (now that the planet is known to be eclipsing, the transit probability is around 7%).

    StevoR:
    Regarding HD 4113’s outer companion, the arXiv paper I mentioned says that a main sequence star can be ruled out because it would have been observed, but either a brown dwarf or a white dwarf star is a possibility. As for additional planets around HD 80606, according to arXiv:0706.1962 the stable zone extends from about 1.75 AU, but with an unstable region at 1.9 AU due to an 8:1 resonance (for comparison, the apastron of HD 80606b is at roughly 0.9 AU). On the other hand, Kozai oscillations are not kind to multi-planet systems: a typical result is eccentricity growth followed by planet-planet scattering. It may well be that HD 80606b survived by virtue of being the most massive planet in the original system, and thus able to eject the other planets it encountered.

    This is not to say that nice near-circular multi-planet systems are impossible in wide binaries: the obvious counter-example to Kozai catastrophe scenario is 55 Cancri, which is a ~1000 AU binary with five known planets around the primary star. It’s also possible to have a system with regular orbits at high inclinations provided that other perturbations such as general relativistic effects or tides, etc. disrupt the Kozai oscillations: there is a fairly obvious example in our own solar system of a regular system of objects in near-circular orbits with an inclination relative to the orbital plane of the parent body well above the Kozai angle…

  69. Also, assuming HD 80606 b is tidally locked (I’m not sure if that was in the report Phil cited or not) then there’s some interesting behavior of the star in the planet’s sky. First, for the days I listed previously the following are the angular diameter and angular speed of the star in the sky:

    Day 1: 0.545 deg, +2.96 deg/day
    Day 7: 0.550 deg, +2.91 deg/day
    Day 24: 0.623 deg, +2.44 deg/day
    Day 30: 0.683 deg, +2.12 deg/day
    Day 34: 0.742 deg, +1.80 deg/day
    Day 37: 0.804 deg, +1.46 deg/day
    Day 45: 1.124 deg, -0.44 deg/day
    Day 48: 1.403 deg, -2.35 deg/day
    Day 54: 5.462 deg, -48.73 deg/day
    Day 55: 16.076 deg, -270.80 deg/day

    Thus, at apastrion HD 80606 appears to be roughly the same size as the sun as seen from the earth, but at periastrion it is 32 times bigger in apparent size. A positive sign for apparent angular speed means east to west motion. At periastrion the star is moving retrograde (i.e., west to east) at the fantastic rate of 270 degrees per day.

    Assume that the planet has zero tilt and that we have an observer at the equator at the sub-stellar point during periastrion. The star is then right over head for this observer at periastrion. The following is a brief description of the star’s motion as viewed by this observer:

    At the start the star (HD 80606) is at the zenith point. It is moving towards the east at an initial rate of 270 degrees per day.

    At time 0.6 days the star sets beneath the eastern horizon.

    At time 4.9 days the star reaches its maximum angle below the easter horizon of 41.4 degrees (of course the observer can’t see it). It then begins to move westward.

    At time 24.0 days the star rises in the east.

    At 39.7 days the star is 45 degrees above the eastern horizon heading westward.

    At 55.0 days the star is back at zenith (this is the apastrion point) and
    heading westward.

    At 70.3 days the star is 45 degrees above the western horizon and heading westward.

    At 86.0 days the star sets in the west.

    At 105.1 days the star reaches its maximum angle beneath the western horizon of 41.4 degrees. It then begins to move eastward.

    At 109.4 days the star rises in the west in moves eastward.

    At 110.0 days the star is back at zenith (periastrion point).

    Thus, in one complete orbital cycle we have the following:

    1.) star sets in the east
    2.) star rises in the east
    3.) star sets in the west
    4.) star rises in the west

    At no point is the star ever below 41.4 degrees below the horizon.

  70. Eccentric orbit leads to pseudosynchronisation rather than synchronisation… the expected rotation period of HD 80606b if the rotation is pseudosynchronised is 40.7 hours. (See oklo.org for the draft of the paper about the HD 80606b Spitzer results).

  71. @andy “Eccentric orbit leads to pseudosynchronisation rather than synchronisation… the expected rotation period of HD 80606b if the rotation is pseudosynchronised is 40.7 hours.”

    Drat once again. Using the data contained in the linked report the only really interesting stuff in terms of star position is the adjustment of the star in the sky relative to where a hypothetical star following a circular orbit would be. We have the following assuming a prograde orbit with a period of 1.72 days:

    Periastrion: star shifted 0 deg
    5 days after periastrion: maximum shift eastward of 130.6 deg
    24 days after periastrion: eastward shift of 90 deg
    55 days after periastrion: star shifted 0 deg
    87 days after periastrion: westward shift of 90 deg
    106 days after periastrion: maximum shift westward of 130.6 deg

    This maximum shift of 130.6 deg is 36% of one rotation period or 15 hours difference between the mean stellar day and the local stellar day.

  72. Asimov Fan

    (I know its a bit late in the comments here but still..) This thought occurred to me the other day whilst thinking of this awesome “planet with a comet’s orbit”; HD 80606 b :

    Now I understand that as andy & Tom Marking noted rings and moons are exceedingly unlikely – even impossible – given HD 80606 b’s orbit but what about trojan type asteroids in a 1:1 resonance similar to those around Jupiter – but perhaps even a trojan the size of Mars or Earth? Or at the L-1 or other Lagrange points?

    Would that be possible? If so, then how massive could it be before the trojan and HD 80606 b’s situation became unstable? Anyone?

  73. StevoR

    @ Asimov Fan : Afraid I don’t really know. Its a neat idea though – sounds great to me! :-)

    @ andy said on Feb. 2nd, 2009 at 1:17 pm :

    “StevoR: Regarding HD 4113’s outer companion, the arXiv paper I mentioned says that a main sequence star can be ruled out because it would have been observed, but either a brown dwarf or a white dwarf star is a possibility. As for additional planets around HD 80606, according to arXiv:0706.1962 the stable zone extends from about 1.75 AU, but with an unstable region at 1.9 AU due to an 8:1 resonance (for comparison, the apastron of HD 80606b is at roughly 0.9 AU). On the other hand, Kozai oscillations are not kind to multi-planet systems: a typical result is eccentricity growth followed by planet-planet scattering. It may well be that HD 80606b survived by virtue of being the most massive planet in the original system, and thus able to eject the other planets it encountered.”

    Thanks for that! Your answer is very much appreciated. :-)

    “Regarding HD 4113’s outer companion, the arXiv paper I mentioned says that a main sequence star can be ruled out because it would have been observed, but either a brown dwarf or a white dwarf star is a possibility.”

    How about another more distant high mass exoplanet? Or wouldn’t it have enough mass to do the trick?

    “It may well be that HD 80606b survived by virtue of being the most massive planet in the original system, and thus able to eject the other planets it encountered.”

    Or did it consume them perhaps? Have seen an awesome doco on migrating exoplanets becoming Hot Jupiter’s and swallowing up the inner rocky worlds -could at least some other protoplanets inthe system have been devoured and incorporated inside the bulk of HD 80 .. Oh can’t we just call it Struve 1341 or something?

    The “Comet Planet” maybe .. Or, dare I suggest, “Velikovsky-Reversed” after the notorious woo purveyor who pseudo-scientifically suggested Venus was ejected from Jupiter as a comet? ;-)

    BTW. My thanks Tom Marking & andy for doing all the maths (my weakest area) & coming up with so much super-luminous (ie beyond just brilliant!) extra info & insight into this wonderful “Comet Planet”! I’m impressed by it – very impressed! ;-)

  74. @Asimov Fan “given HD 80606 b’s orbit but what about trojan type asteroids in a 1:1 resonance similar to those around Jupiter – but perhaps even a trojan the size of Mars or Earth? Or at the L-1 or other Lagrange points?”

    I believe the Lagrange points are derived assuming a circular orbit or a near circular orbit. Maybe Andy can correct me if I am wrong there. So they wouldn’t apply to such an elliptical orbit. There would be no L4 or L5 point in the same path of the orbit. At least that’s my understanding.

  75. The question is, is “Struve 1341Bb” really so much better than HD 80606b? The latter at least has the advantage of being more commonly used, so at least some minuscule fraction of the population will know what you’re talking about, rather than confusing everybody…

    As for 1:1 resonances in eccentric systems… I have no idea. I suggest you go download a copy of Gravity Simulator or something similar and play with it yourself. There are other possible 1:1 resonant configurations than Trojans (tadpole orbits), e.g. quasi-satellites and horseshoe orbits, which may or may not exist as possibilities in the HD 80606 system. And once you start letting the system become non-coplanar, things have the potential to get really weird. And apparently the Kozai mechanism can theoretically lead to a system becoming captured in a retrograde resonance, so don’t feel limited to having your simulated planets going in the same direction around the star…

  76. Asimov Fan

    @StevoR : Thanks!

    Tom Marking said :

    “@Asimov Fan “given HD 80606 b’s orbit but what about trojan type asteroids in a 1:1 resonance similar to those around Jupiter – but perhaps even a trojan the size of Mars or Earth? Or at the L-1 or other Lagrange points?”

    I believe the Lagrange points are derived assuming a circular orbit or a near circular orbit. Maybe Andy can correct me if I am wrong there. So they wouldn’t apply to such an elliptical orbit. There would be no L4 or L5 point in the same path of the orbit. At least that’s my understanding.”

    Thanks for the bad news. I guess I should have realised Lagrange points only apply to normal orbits. :-(

    & andy said :

    “The question is, is “Struve 1341Bb” really so much better than HD 80606b? The latter at least has the advantage of being more commonly used, so at least some minuscule fraction of the population will know what you’re talking about, rather than confusing everybody…

    As for 1:1 resonances in eccentric systems… I have no idea. I suggest you go download a copy of Gravity Simulator or something similar and play with it yourself. There are other possible 1:1 resonant configurations than Trojans (tadpole orbits), e.g. quasi-satellites and horseshoe orbits, which may or may not exist as possibilities in the HD 80606 system. And once you start letting the system become non-coplanar, things have the potential to get really weird. And apparently the Kozai mechanism can theoretically lead to a system becoming captured in a retrograde resonance, so don’t feel limited to having your simulated planets going in the same direction around the star…”

    Okay, maybe not so bad news – just really bizarre! I’m trying to get my head around that and struggling – one planet prograde, another retrograde and no head-on collisions? How does that work?

    Tadpole orbits – are these the same as horseshoe ones? Like Earth’s quasi-moons mentioned by Phil Plait ages ago?

    Gravity Simulator program? Where do I get a hold of that from & how much is it? Are there computer limitations as to which machines will run it?

    As for “Struve 1341Bb” really so much better than HD 80606b?” as suggested by StevoR; I agree with your point there – although, on the other hand, Struve 1341b is a smidgin shorter and at least makes a proper word rather than being just initials and numerals. I do agree with StevoR that it’d be nice if the IAU named at least some of the more notable exoplanets with easy to pronounce and popularise common names.

    Of course, the problem comes when you consider that there’s already over 300 exoplanets now known and new ones are being found – and breaking records – all the time.

    Btw. I second StevoR’s thanks to Tom Marking & andy. Nice work both of you. :-)

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 »