≡ Menu

White Dwarfs and Habitable Planets

Before I get into today’s story, which is an interesting study on planets around white dwarfs that Andrew Tribick passed along, I want to say a few words about Japan. Centauri Dreams has many, many readers in that country, and the terrible images and stories coming out of there have haunted me these past few days. The suffering of those displaced by the earthquake and tsunami, and the continued problems in resolving the worsening situation at Fukushima, make it hard to focus on any other topic. Speaking for myself here at Centauri Dreams — and I know I speak for the entire Tau Zero Foundation as well — you Japanese readers remain in our thoughts and prayers, and will continue to do so until these great national wounds are healed.

On the space front, today is the day when MESSENGER enters Mercury orbit. Below is the schedule for the events, which we’ll follow closely as orbital insertion occurs.

White Dwarfs and Potential Planets

But for now let’s talk about white dwarfs, those interesting survivors of Sun-like stars that have gone through the red giant phase and presumably swallowed up planets within roughly Earth’s distance from the Sun. An interesting paper from Eric Agol (University of Washington) takes a look at exoplanet possibilities around white dwarfs, and draws some surprising conclusions. We have, of course, searched for habitable planets primarily around stars that are much younger, assuming that a planetary system that had undergone the transformation of a red giant into a white dwarf would be unlikely to provide a suitable home for life. But Agol isn’t so sure.

Remember the process: Stars like the Sun eventually exhaust their nuclear fuel and at some point lose their outer envelope, leaving only the hot core behind. The core, now a hot white dwarf at temperatures exceeding 100,000 Kelvin, will begin a long process of cooling. A typical white dwarf might be half as massive as the Sun, but not much larger than the Earth in size, and as this NASA article points out, that means it’s extremely dense, perhaps 200,000 times as dense as the Earth itself. When it comes to matter, only neutron stars surpass that density.

Agol points out that the most common white dwarfs have surface temperatures in the range of 5000 K, which leads to his calculation that a planet would need to orbit no closer than about 0.01 AU to be at a temperature where liquid water could exist on the surface. What’s intriguing from the standpoint of finding such planets is that a potentially habitable world like this, Earth-sized or even smaller, would in principle be detectable because of the small size of the host star. The white dwarf, in fact, could be completely eclipsed by a habitable planet that orbits it.

But how does a planet survive the preceding red giant phase? One possibility is that new planets could form out of gases near the white dwarf, especially in binary systems where gravitational interactions could play a helpful role. We know of two neutron stars that have planets that conceivably formed from the disk created after a supernova event. Moreover, the pulsar 4U 0142+61 has been shown to have a circumstellar disk thought to have been formed from supernova debris. Planetary capture or migration can’t be ruled out, either.

Defining a Habitable Zone

I’m going to post Agol’s chart on white dwarf habitable zones (WDHZ) below to illuminate what he has to say. Here the habitable zone is plotted against time as a blue-shaded region, and because the white dwarf is cooling, the region shrinks with time. The planet starts off too hot for liquid water, passes through the white dwarf habitable zone, and then becomes too cold for life.

Image: The WDHZ for MWD = 0.6M⊙ vs. white dwarf age and planet orbital distance. Blue region denotes the WDHZ. Dashed line is Roche limit for Earth-density planets. Planets to right of dotted line are in the WDHZ for less than 3 Gyr. Planet orbital period is indicated on the top axis; white dwarf effective temperature on the right axis. Luminosity of the white dwarf at different ages are indicated on right. Credit: Eric Agol.

Using the WDHZ limits, Agol defines a ‘continuously habitable zone’ (CHZ) as a range of orbital distances habitable for a minimum duration. Choosing a minimum duration of 3 billion years produces a continuous habitable zone within 0.02 AU, so we have a three billion year period for the development of life at that distance. The author comments on the consequences:

…the range of white dwarf temperatures in the portion of the CHZ within the WDHZ is that of cool white dwarfs, ≈ 3000–9000 K (right hand axis in Fig. 1), similar to the Sun. At the hotter end higher ultraviolet flux might affect the retention of an atmosphere, these planets would need to form a secondary atmosphere, as occurred on Earth. Excluding higher temperature white dwarfs only slightly modifies the CHZ since they spend little time at high temperature. Cool white dwarfs are photometrically stable…, which is critical for finding planets around them.

Finding a White Dwarf Planet

An interesting prospect indeed, one that Agol further explores by simulating sky surveys that could find such planets. Among the latter calculations, it’s interesting to note that the GAIA mission will observe 200,000 disk white dwarfs between 50 and 100 times each, making the detection of a white dwarf with a habitable planet a real possibility. Even more likely are the prospects for the Large Synoptic Survey Telescope, a planned wide-field survey in Chile.

And what would life be like on a planet orbiting in the habitable zone of a white dwarf?

The most common white dwarf has Teff [effective temperature] ≈ 5000 K, close to that of the Sun; consequently, inhabitants of a planet in the CHZ will see their star as a similar angular size and color as we see our Sun. The orbital and spin period of planets in the CHZ are similar to a day, causing Coriolis and thermal forces similar to Earth. The night sides of these planets will be warmed by advection of heat from their day sides if a cold-trap is avoided… Transit probabilities of habitable planets are similar for cool white dwarfs and Sun-like stars, but the white dwarf planets can be found using ground-based telescopes… at a much less expensive price than space-based planet-survey telescopes.

This is a provocative paper, one that jolts us into thinking about habitable zones in places where we hadn’t thought of looking before. Yet as we’re finding in our exoplanet research, the universe keeps yielding surprises, and a habitable planet around a white dwarf may not be so bizarre after all. Does anyone know of any science fiction writers who have tackled such a scenario? If so, do let me know. Agol’s paper is “Transit Surveys for Earths in the Habitable Zones of White Dwarfs,” in press at the Astrophysical Journal Letters and available as a preprint.

Addendum: See the comments below for a link to a discussion of white dwarf planets that I was hitherto unaware of.


Comments on this entry are closed.

  • James Davis Nicoll March 17, 2011, 12:30

    This is a provocative paper, one that jolts us into thinking about habitable zones in places where we hadn’t thought of looking before.


    I think I eventually thought to look at the orbital velocity of Azure: it’s something like 135 km/s, which means incoming meteors will income with a lot more energy than on Earth.

  • McNeil Gaila March 17, 2011, 12:40

    But he didn’t tell how that planet survived from the death of red giant…could it be possible that the planet encircling the white dwarf forms from the escaped gases?

  • Randy McDonald March 17, 2011, 12:55

    I know that Douglas Muir discussed the subject some time ago in the Usenet group rec.arts.sf.science, and the subject was later raised at sci.physics here:


  • ant6n March 17, 2011, 13:57

    A planet could probably survive if it’s far enough initially from the host star – of course then it’d be much farther away than 0.2 AU. But maybe it’s orbit could decay from the safe distance to 0.2 AU due to drag against the matter that the star is loosing.

  • Randy McDonald March 17, 2011, 14:58

    Perhaps the core of an epistellar Jovian that survived? Would such a stripped core be more likely to be a superterrestrial world?

  • Michael March 17, 2011, 16:00

    Would’nt the gravity gradient on the planet be a problem as it will only be less than a million miles from the stars surface and it will probably be tidally locked?

  • ljk March 17, 2011, 16:16

    Excellent sentiments, Paul. Many people (too many being even one) feel that somehow Earth and humanity are disconnected and special from the rest of the Universe and that any serious focusing on it via space exploration, especially in times of crises like now, is both irrelevant and even disrespectful. How little they understand that not only is the truth of it all the opposite of what they think, but that the grander perspective can even help and heal over the long run. Plus the ultimate tragedy regarding events like the one in Japan is if we let them defeat us as a species and society and take away all the good we have accomplished so far.

    Case in point: I was watching the PBS Newshour last night (refreshing compared to the contemporary nightly news from other television networks; its no-rush pace and serious tone reminded me of network news from decades ago). Naturally they led off and occupied most of the hour with the natural and artificial disasters befalling Japan. However, the last segment was about MESSENGER’s impending destiny with Mercury. This was framed by Miles O’Brien doing a tour of an exhibit at the National Air and Space Museum in Washington, DC, showcasing beautiful images of various worlds in our Sol system taken by space probes from the book Beyond. It really put things in their proper perspective as well as offering a bit of relief from and hope for the latest tragedy on Earth.

  • Mike March 17, 2011, 18:42

    While I don’t see any reason why we shouldn’t look for planets hosted by White Dwarfs I think it would be a very harsh stellar enviroment for life.
    The thermal habitable zone is right on the WDs doorstep. White Dwarfs can have extremely powerful magnetic fields with all that implies for radiation and there is the question of tidal effects experienced by close-in HZ planets.
    Still, I really like the audacity of the idea and agree that WD exoplanet surveys should be conducted. As for my questions about the suitability of WD hosted planets for life? I remind myself we are working from a biological
    sample of one. What further surprises the Universe has waiting for our curious minds? Maybe something could be living on those planets.

    That was a wonderful and heartfelt sentiment you wrote at the begining of this article regarding Japan’s troubles Paul. Well said.

  • Paul Gilster March 17, 2011, 20:33

    Thank you, Mike, re the sentiments about Japan. I appreciate your support. That goes for ljk, too — thanks to both of you.

  • Enzo March 17, 2011, 21:04

    Very interesting planets in a very deep gravitational well.
    Unless I made a mistake, the escape velocity from 0.01 AU away from a WD 0.6 solar masses is some 325 Km/s.

  • Adam March 18, 2011, 1:32

    Hi All
    Larry Niven’s “Inconstant Moon” anthology featured a story on Earth after it had become tide-locked and orbiting the white dwarf Sun – the temporal castaways had to spin Earth up again to defrost the atmosphere.

  • Stevo Darkly March 18, 2011, 13:33

    ^ I believe the title of that particular story was “One Face.”

  • andy March 18, 2011, 14:57

    Would’nt the gravity gradient on the planet be a problem as it will only be less than a million miles from the stars surface and it will probably be tidally locked?

    Yes a planet in the habitable zone of a white dwarf would be expected to be tidally locked. That would make them valuable tests for models of planetary climate: you get to have a synchronously-rotating planet around a star with similar temperature to the sun (thus a similar spectral energy distribution, unlike that of a red dwarf star), and a similar rotation rate to the Earth. Probably such worlds are rare, but finding one would be an exceptionally interesting discovery.

  • bigdan201 March 18, 2011, 15:26

    Indeed, condolences to Japan. They have contributed a great deal to the world – for me personally, Nintendo played a large role in my childhood.

    As far as exoplanets, I believe that they will continue to surprise us. The more we search, the more we find. After all, who expected to find pulsar planets?

  • Rob Henry March 18, 2011, 21:10

    Talking about the large tidal forces and how white dwarfs could have close-in planets, why the need to hypothesise that some planets could be drawn in by gas drag from the red giant phase when we would otherwise expect this process to be too rapid?

    I remember reading (possibly from these pages) that tidal interactions between a red giant’s outer layers and its planets cause them to spiral in anyway. Has that proposed mechanism been disproven?

  • Eniac March 20, 2011, 0:22

    It seems to me that for every planet spiraling in too close, there could be another further out spiraling in just enough to take its place.

  • Ronald March 21, 2011, 7:56

    Form a statistical viewpoint, how relevant are these WDs really?

    I mean two things:

    – first of all: how abundant are WDs in our galactic disc? My impression is not too common, although observational bias may also play a role here. But anyway probably much less common than red dwarves (M).

    – Secondly: from the diagram we can see that the widest HZ, particularly of the most common WDs, is only about 0.01 AU wide. This is very relevant with regard to the statistical probability that a suitable planet will be situated within this HZ. This HZ width is very, very narrow in comparison with that of a typical solar type star (easily from o.20 – 0.50 AU) and even in comparison with the HZ of a ‘typical’ red dwarf (0.02 – 0.05 AU)

    Combining these two factors, low abundance of WDs and very narrow HZ, results in a very small total potential real estate for WD terrestrial planets.
    This in combination with all the above-mentioned downsides of WDs leaves me wondering: how (ir)relevant are the WDs really with regard to habitable planets?

  • andy March 21, 2011, 16:38

    @Ronald: IIRC most white dwarfs are the remnants of A- and B-type stars, these aren’t particularly common as these things go but not particularly uncommon either.

    The absolute width of the habitable zone is probably less relevant than you think. Planetary distribution is closer to being logarithmic than linear: the planets near the star are bunched closer together. This works to counteract the small absolute width of the habitable zone. Taking red dwarf planetary systems as an example (for which the narrow habitable zone argument has also been used), Gliese 876 has two planets in its habitable zone, and Gliese 581 may have 1 or 2 depending on whether planet g is confirmed. This is despite having much narrower habitable zones than the Sun does: on a logarithmic scale there isn’t all that much difference.

    Of course there are other issues regarding whether it is possible to get terrestrial planets that close-in to a white dwarf in the first place (I’d guess such systems are going to be pretty rare in the grand scheme of things), but if you can then hitting the habitable zone shouldn’t be too big a problem.

  • Rob Henry March 21, 2011, 17:59

    Ronald I think your analysis misses one crucial point: testability. WD may look like poor candidates for life to most of us but the very constrictions that apply to allow thriving ecosystems, also allow rapid and easy testing of the hypothesis. We could soon find the true proportions of WD that harbour suitable planets in their HZ if we ever thought that a worthy task.

  • Brian Helm March 24, 2011, 18:16

    The probability that anyone who reads this blog would know or should know of Larry Niven’s The Integral Trees and sequel The Smoke Ring must be close to a 100%.These wonderful novels describe what life would be like around a white dwarf. as with most of his novels he tries to keep the science as accurate as the story requires.

  • Paul Gilster March 24, 2011, 18:33

    Thank you, Brian! Hadn’t realized Larry Niven had explored this setting. I’ll look into these — I’ve heard of both titles but haven’t read them yet.

  • Rob Henry March 25, 2011, 17:12

    Brian, Niven’s novel is actually set around a neutron that emits an atypically low level of hard radiation. The planet is just outside its Roche limit for a body of its density when its solid core is considered, and just inside it when the whole planet (including its atmosphere) is considered. This spreads that Jovian’s atmosphere in a ring around its orbit. This ring would dissipate, were it not for the tight orbit allowed around a neutron star.

    Thus you change of setting bring up an interesting question of whether orbits around very old white dwarfs could be tight enough to allow this setup without the requirement of finding a neutron star with such ridiculously low levels of hard radiation emission. If so such smoke ring worlds look far more possible than Niven could have ever imagined.

  • Brian Helm March 26, 2011, 15:26

    Glad to be of service Paul. I really enjoy reading your blog i have no actual education in science yet i understand more science and physics than your average guy. If you like good science in your books try Robert Forwards Chella novels Dragons Egg and Starquake. They are about how far “Habitability” can be taken. also his Flight of the Dragonfly and associated rocheworld novels are excellent reads.