≡ Menu

Dying Stars and their Planets

Although I can’t make the journey just then, I wish I could attend an upcoming conference at Arecibo (Puerto Rico) called Planets around Stellar Remnants. The meeting takes place twenty years after the discovery of the first exoplanets, the worlds orbiting the millisecond pulsar PSR B1257+12. I’ve been interested in the fate of planets around older stars ever since reading H.G. Wells’ The Time Machine as a boy and encountering his image of a swollen, dying Sun. We also have interesting questions to ask about the kind of planets that exist around white dwarfs, and whether new planets (and chances for life) may eventually occur in their systems.

It’s appropriate that the conference be organized by Penn State, for it was that university’s Alexander Wolszczan, working with Dale Frail of the National Radio Astronomy Observatory, who made the discovery of those two and possibly three planets that launched the modern exoplanet era. Nor has Wolszczan slowed in his efforts. The most recent work out of Penn State is the discovery of planets around three dying stars — HD 240237, BD +48 738, and HD 96127 — one of which has an interesting and still unidentified object, perhaps a brown dwarf, in orbit around it.

All three of the stars are much further along in their lifelines than our Sun, says Wolszczan:

“Each of the three stars is swelling and has already become a red giant — a dying star that soon will gobble up any planet that happens to be orbiting too close to it. While we certainly can expect a similar fate for our own Sun, which eventually will become a red giant and possibly will consume our Earth, we won’t have to worry about it happening for another 5 billion years.”

What we get by studying highly evolved stars like these is an updated window into planet formation. 30 known planets and brown dwarf-mass companions are now known to exist around giant stars (all three stars studied here are K-class giants). The object around BD +48 738 is tricky because a long-term radial velocity trend here indicates a distant companion but the data are not yet sufficient to decide between a planet and a low-mass star as the culprit. The star is also orbited by a planet of about 90 percent Jupiter’s mass at roughly 1 AU in a 400-day orbit.

I’ll pause on this because we’re finding a number of companions to giant stars that have minimum masses of about 10 Jupiter masses, making them either brown dwarf candidates or massive planets. It will take further work to identify the object in the outer system of BD +48 738, but if it does turn out to be a brown dwarf, then we have another case of a system with a Jupiter-class inner planet and a distant brown dwarf orbiting the same star. The paper on this work discusses the implications in terms of our primary theories of planet formation:

In principle, such a system could form from a sufficiently massive protoplanetary disk by means of the standard core accretion mechanism (Ida & Lin 2004), with the outer companion having more time than the inner one to accumulate a brown dwarf like mass. A more exotic scenario could be envisioned, in which the inner planet forms in the standard manner, while the outer companion arises from a gravitational instability in the circumstellar disk at the time of the star formation (e.g. Kraus et al. 2011). In any case, it is quite clear that this detection, together with the other ones mentioned above, further emphasizes the possibility that a clear distinction between giant planets and brown dwarfs may be difficult to make…

The three stars appear jittery under observation because they oscillate more than younger stars like the Sun. That made the planet hunt a challenge, but also allowed for the discovery of a negative correlation between the star’s metallicity and the degree to which it oscillates. Wolszczan says that the less metal content the team found in each star, the more noisy and jittery it turned out to be. The paper relates this to p-mode (pressure-driven) oscillations at the surface of the star:

The origin of this trend is most likely related to the fact that higher metallicity (opacity) of the star lowers its temperature, which decreases the amplitude of p-mode oscillations, while lower metallicity has the opposite effect.

This Penn state news release quotes Wolszczan on the future of our own Solar System as the Sun swells to become a red giant and swallows the inner planets. Somewhere in the remote future, says the astronomer, perhaps one to three billion years from now, we may consider moving to Europa, an icy wasteland that under the gaze of a swollen Sun will become a world of ‘vast, beautiful oceans.’ It’s an enchanting thought, and one we’ll doubtless think more about as we continue our investigations of giant stars and the fate of planetary systems around them.

The paper is Gettel et al., “Substellar-Mass Companions to the K-Giants HD 240237, BD +48 738 and HD 96127,” accepted by the Astrophysical Journal (preprint).


Comments on this entry are closed.

  • Interstellar Bill November 3, 2011, 13:01

    Good luck on a planet with no land.

    It was recently shown that a 100-mile asteroid
    passing in front of the Earth every 10,000 years
    could move Earth outward at a constant irradiance.

    This would be a lot easier than moving humanity to Raftworld Europa.

  • andy November 3, 2011, 18:56

    I’ve been interested in this idea of Europa becoming an ocean world under the red giant sun for a while. The escape velocity is discouragingly low, I guess the question is how the atmospheric escape and the evaporation of the water reservoir would balance out. Anyone aware of any research that considers what the conditions on Europa under the red giant sun would be like?

    As for moving the Earth outwards, good luck getting the other planets out of the way, avoiding various destructive resonances and all that… and of course correcting for the ageing of the planet itself as the tectonics become less active, the core cools off, etc…

  • Daniel Suggs November 3, 2011, 23:40

    By the time this happens, we should be a totally space faring race, if we have survived. And, if we even know where our planet of origin is, we may want to save it for historical reasons.
    Okay, wishful thinking, but possible.

  • Doug M. November 4, 2011, 4:17

    The Europa thing is a nice image, but it has no basis in reality.

    The Sun will take about 6 billion years to finish climbing the main sequence, at which point it will “only” be about twice as luminous as it is today. That’ll be enough to render Earth uninhabitable — but not remotely enough to thaw Europa, which will remain a frozen iceworld. Surface temperatures during the day at low latitudes are around 120K; in 6 billion years they’ll rise to around 150K. An improvement, but still 100+ degrees below the melting point of water.

    When the Sun becomes a red giant, though… well, it’s going to be around 2000 times as luminous as today. The habitable zone will be out near what’s now the Kuiper Belt. Europa’s surface temperature will be around 800K or over 500C — hotter than contemporary Mercury. The ice will have long since boiled away, and nothing will remain of Europa but a small rocky core.

    Of course, even if Europa *did* get the precisely right amount of insolation, there’d still be no seas, because Europa is much too small to hang on to an atmosphere.

    Europa is less massive than the Moon, people. Do you imagine “vast, beautiful oceans” on the Moon? No? Then don’t go putting them on Europa.

    Doug M.

  • Adam November 4, 2011, 4:44

    The timespan quoted is a bit short. Europa won’t become habitable until the Sun is climbing the Red Giant Branch over 7.5 billion years from now. Its habitable time was estimated in an article from a few years ago…

    Solar Evolution and the Distant Future of Earth

    …as being between a Solar age of 12.07-12.10 billion years, thus 30 million years of clement weather. Might not be long enough to worry about atmospheric erosion via Jeans Escape.

  • Ole Burde November 4, 2011, 7:48

    It seems a bit far out to worry about something 500 mil years into the future , when other kinds of disasters piles up inside the 100 year horizont , like anny kind of progress beeing crushed by the impossible population load of 15 bill people at around year 2100 .

  • ljk November 4, 2011, 13:30

    Dying suns is one of several places SETI should be looking at. The reasons range from transmissions from occupying civilizations asking for help or sending their collective knowledge and history into the galaxy for preservation (see James Gunn’s SF novel, The Listeners, for the latter).

    For those species which can escape their dying solar systems, SETI should look for the signatures of propulsion units pushing starships away from their suns. Or perhaps they should also watch for ships heading towards such systems, either as part of a rescue effort or part of a plundering scheme to grab whatever is left. What, you think intelligent beings won’t be greedy and materialistic just because they have starships?

    I am surprised that Titan has not gotten a menti0n here in the comments section, as that giant moon has also been described as an Earth frozen in its early stages. Well a big red sun might just thaw things out for a while, perhaps enough to at least get something organic started before it shrinks to a white dwarf and plunges everything into a deep, deep eonical freeze.

    But what I am *really* surprised at is the assumption that in the next few billion years the Sol system won’t be turned into something more interesting and useful to a really advanced species, or that cosmic forces such as a passing star won’t disrupt things by then. And remember, the Andromeda galaxy and the Milky Way are going to collide in just a few billion years time, which might really play billiards with the Children of the Sun.

  • Adam November 4, 2011, 16:36

    Larry, Greg Matloff has discussed that very issue in a couple of papers available at his personal website…

    His SETI page

    Red Giants and Solar Sails

    …he suggests searching red dwarfs orbiting white dwarfs for signs of civilisations that have moved.

  • Adam November 4, 2011, 16:58

    Doug M, the Moon can retain an atmosphere. What it probably can’t do is retain it against aeons of solar wind erosion.

    However when the Sun is climbing the Red Giant Branch, the solar wind speed will be declining, thus becoming less effective at eroding atmospheres. The climb to Red Giant luminosity takes about a billion years after the relative constancy of the Redwards Traverse. The luminosity maximum is achieved in the last few million years, thus is only a very short period compared to the time at lower luminosities.

    Best thing about Red Giants for small bodies is that the UV levels go way down, meaning the exosphere isn’t as bloated from UV absorption. Coupled with a slower solar wind and Europa should hold on to its atmosphere while it’s habitable. Titan becomes interesting long before Europa because of its methane reservoir and probable ammonia. It will defrost about the same time, but thanks to the ammonia/water oceans it should last about twice as long. Ralph Lorenz, Jonathan Lunine & Chris McKay wrote the now classic paper on it…

    Titan under a Red Giant Sun

    …though “habitable” is a matter of taste when talking of ammonia/water oceans. If our descendants last 7.5 billion years, then they’ll find such an environment fairly straightforward to tweak a biosphere for. Europa at least will defrost with an oxygen rich atmosphere. If there’s any CO2 mixed in the ice, then Europa might begin defrosting long before water starts sublimating in earnest.

  • LarryD November 4, 2011, 18:08

    In 500 million years we’ll need that planet moving trick, because Sol’s luminosity will have increased enough to move the HZ outward past Earth’s current orbit. Or so I’ve read.

    I understand that out galaxy is already the result of several galactic “collisions”, and our solar system has survived, even keeping it’s Oort cloud.

  • andy November 5, 2011, 4:15

    Regarding red giant stellar winds, what is the composition of these? As I understand it the lower surface gravity of the red giant star prevents it from retaining a hot corona, so I’d expect the red giant wind to have a greater proportion of neutral particles. This would surely reduce the effectiveness of magnetic fields as a protection mechanism.

    And while the velocity is indeed lower, the outflow rate is substantially greater, which effect would win out in terms of the efficiency of erosion of planetary atmospheres?

  • Adam Crowl November 5, 2011, 4:41

    Hi ljk & LarryD,
    LJK, I forgot to add that the collision with M31 won’t be likely to result in ANY stellar collisions, though plenty of perturbations are likely, but stars getting even as close as the Oort Cloud is pretty rare and even rarer is getting as close as the Kuiper Belt. What will happen is lots and lots of star-making which could result in a bunch of supernovae!

    LarryD, the “500 million years” before Earth overheats is based on a certain model of Earth’s response to the Sun’s gradual rise in luminosity. That particular model assumes everything else will remain the same, but that’s unlikely. If the partial pressure of nitrogen declines, then the greenhouse effect from carbon dioxide will decline and the Earth could remain habitable to life for another 2.3 billion years. Alternatively the greenhouse instability of the Earth is driven largely by the thermal response of the oceans – if Earth became a desert planet, then it would remain habitable until the Sun reaches ~1.7 times its present output. Combined with a reduced atmospheric pressure, it means Earth might remain habitable until the end of the Sun’s Main Sequence in 5.5 billion years.

    But this all assumes no technological intervention. Several scenarios are possible – a variably reflective shell engulfing the Earth is the simplest. Planet moving and Solar engineering are more dramatic possibilities. Given sufficient thrust a leisurely spiral of the Earth outwards from the Sun would compensate for the brightening, though the pace of travel would need to be rather rapid for a 6 billion trillion ton planet to sustain to escape the more dramatic stages of the Sun’s Red Giant Branch (RGB).

    Once the Sun hits the Horizontal Branch/Helium Main Sequence, the habitable zone will be roughly where Jupiter will be – the Sun’s mass loss during the RGB will cause all the orbits to expand by ~30%. The HB offers just 110 million years of stability before the Sun begins a series of dying spasms known as the Asymptotic Giant Branch. Not healthy for any of the planets. If the RGB’s mass-loss can be tweaked a bit, then the Sun won’t hit the HB at all and will slowly decline into being a helium white dwarf. Earth can remain in its habitable zone then for billions more years, if it spirals inwards as it cools.

  • Rob Henry November 5, 2011, 19:53

    Andy, you’ve got me interested. If there really are so many neutral particles in that reduced velocity red giant wind then wouldn’t we have to do many calculations before we could know if their net effect was to add to Europa’s nascent atmosphere or oblate it?

  • Doug M. November 6, 2011, 2:41

    Adam, I’d like to see a cite for the proposition that the Moon could maintain a breathable atmosphere at STP for geological periods of time.

    Meanwhile, note that Europa sits deep inside Jupiter’s radiation belts. Radiation sputtering is believed to be the reason Ganymede has no atmosphere. Europa gets about an order of magnitude more energetic particle impacts than Ganymede. So reddening of the Sun wouldn’t help, since sunlight would probably not be the major driver of atmospheric loss.

    Doug M.

  • Rob Henry November 6, 2011, 16:17

    Doug M, it is my understanding that the prime driver of the high level of charged ion radiation on the inner three Galileans is the volcanic activity on Io. How will this be effected over the next 5 billion years as Jovian tidal bulges act to push the orbits of these moons outwards?

  • Adam Crowl November 7, 2011, 3:49

    Doug M

    Crack open a planetology textbook and read up. Mine is “Planetary Science” by George Cole & Michael Woolfson. Topic O.2.2, pp297-301.

    As for the Galileans, odds are they lost their initial atmospheres from overly energetic impactors. Being around a lighter primary and further out, meant Titan retained its. Solar-wind sputtering is unlikely to have been the cause.

  • Doug M. November 7, 2011, 5:09

    Rob, that’s a reasonable question. Io is indeed migrating outward. However, it’s a very, very slow process — by the time the Sun becomes a red giant, Io’s semimajor axis will only have increased by about 20%. See


    — pages 82 through 87. So the radiation environment at Europa will be nearly as bad as it is today.

    Also, I note in passing that Europa has no magnetic field. (Which is a bit of a mystery, since it ought to have a small one.) So its hypothetical atmosphere would be directly exposed to particle bombardment from both Jupiter and the Sun.

    If you’re absolutely determined to find shelter on a moon of Jupiter… well, I don’t fancy your chances, but you’re better off looking at Ganymede or Callisto. They’re both much more massive than Europa (Ganymede is more massive than Titan, about a quarter as massive as Mars), they’re out of the worst of the radiation (though Ganymede still gets a lot) and Ganymede even has a magnetic field.

    People overfocus on Europa. (Because of the whole “woo Europa has an ocean and OCEANS MEANS LIFE!” thing, of course. Which I suspect is one day going to look as silly as putting naked barbarian princesses on Mars, but that’s a comment for another day.)

    Doug M.

  • Doug M. November 7, 2011, 11:47

    Adam, I’m sorry, but the Cole & Woolfson is not available online. Can you give us a cite that is? I’m familiar with the rule-of-thumb that if average molecular velocity > escape velocity/10, no atmosphere — but if you can provide more detail, I’d be very interested.

    This is the first time I’ve heard the theory that “odds are [the Galileans] lost their initial atmospheres from overly energetic impactors.” AFAIK current thinking is that Ganymede lost its primordial atmosphere to sputtering driven by Jupiter’s magnetosphere. See, e.g., Johnson, R.E., Magnetospheric-Plasma-Driven Evolution of Satellite Atmospheres, available online at http://caps.space.swri.edu/caps/publications/CorrectedTitan.pdf.

    “[It] is shown that if after 0.5 Gyr any of the Galilean satellites had retained or acquired a Titan-like atmosphere, Io and Europa would have lost it due to the interaction with the Jovian magnetosphere even at present atmospheric escape rates. However, larger magnetospheric plasma densities would have been required early in Jupiter’s history for Ganymede and Callisto to have lost such an atmosphere.”

    — In other papers, Johnson and others have speculated that if the Galileans had ever possessed significant atmospheres, they would have provoked a feedback interaction with the magnetosphere — viz., as they got stripped away, they’d contribute more ions to the particle environment, which in turn would accelerate the stripping process. Basically, the atmospheres themselves would have created that “larger magnetospheric plasma density” required above. Speculation, but interesting.

    Googling various combinations of “Galilean”, “atmosphere” and “impactors” gives me nothing, though. Is this theory somewhat obscure, or are my google skills just failing me?

    Doug M.

  • andy November 7, 2011, 16:45

    This brings up the question of what is likely to happen to the radiation environment around the gas giants as the solar wind’s characteristics change with the evolution of the Sun.

    Any predictions of the rotation rate change for the gas giants themselves, I would guess they ought to be slowly spinning up due to gravitational contraction, as is the case for brown dwarfs which lack the magnetic braking mechanisms that operate in late-type stars.

  • Adam November 8, 2011, 3:32

    Hi DougM.
    Here’s the paper…
    Impact-generated atmospheres over Titan, Ganymede, and Callisto
    …by Zahnle, Pollack, Grinspoon & Dones. Older paper. Thanks for the link – I will read it with interest.

  • Adam November 8, 2011, 4:05

    Hi DougM.
    Reading that paper implies strongly that Europa would retain its atmosphere for the 30 million years it will be habitable during the climb up the Sun’s RGB. If it loses a Titan worth of atmosphere in 4 Gyr, as the paper states, then the loss over just 30 Myr is minor. That and the slower solar wind during the RGB would mean a sufficiently long-lived atmosphere.

  • Doug M. November 13, 2011, 16:04

    Hi Adam,

    I got engaged by this impacts vs. sputtering question, and ended up sending e-mails to several of the article-writers, including R.E. Johnson and David Catling. Both were gracious enough to respond at a little length.

    Apparently this is an ongoing conflict! — Okay, a very polite, low-key conflict. But apparently impacts and sputtering are both perfectly plausible models for stripping atmosphere off the Galileans . And apparently we do not have, at the moment, compelling evidence as to whether the correct answer is either, neither, or both. Both models are supported by a fair amount of indirect evidence, but neither can be conclusively affirmed at this time. And “both are right” is actually a possible answer; there’s a widely admired recent paper that suggests that Mars started with a dense atmosphere, lost most of it to impacts, and then lost most of what was left to solar sputtering.

    A fascinating side trek; thanks for getting me started.


    Doug M.

  • Adam November 17, 2011, 7:26

    Thanks for chasing that issue DougM. Bruce Jakosky wrote a paper which eroded away an original dense atmosphere on Mars by that very means – impacts, sputtering and finally adsorption into all that regolith from the impacts. Seems more likely than a single cause process.

  • ljk May 4, 2012, 9:46

    Will This Be The Fate Of The Earth?

    by Jason Major on May 3, 2012

    Astronomers have found four nearby white dwarf stars surrounded by disks of material that could be the remains of rocky planets much like Earth — and one star in particular appears to be in the act of swallowing up what’s left of an Earthlike planet’s core.

    The research, announced today by the Royal Astronomical Society, gives a chilling look at the eventual fate that may await our own planet.

    Astronomers from the University of Warwick used Hubble to identify the composition of four white dwarfs’ atmospheres, found during a survey of over 80 such stars located within 100 light-years of the Sun.

    What they found was a majority of the material was composed of elements found in our own Solar System: oxygen, magnesium, silicon and iron. Together these elements make up 93% of our planet.

    In addition, a curiously low ratio of carbon was identified, indicating that rocky planets were at one time in orbit around the stars.

    Full article here: