≡ Menu

Habitable Worlds and Hot Jupiters

What happens to potentially habitable planets when a gas giant swings through the neighborhood? It’s a pertinent question when you consider the surprises that ‘hot Jupiters’ have given us. 22 percent of known extrasolar planets show an orbital radius of less than 0.1 AU, and 16 percent are located within 0.05 AU of their host star. That’s a surprise given the assumption that these gas giant planets must form much further out in their systems, but it can be explained by inward migration of the giant planet, a process under much study that is generally thought to be caused by interactions with the protoplanetary disk.

A hot Jupiter seared by its star

Such a migration would seem to spell trouble for planets already orbiting closer to the star, leading some to believe that systems with hot Jupiters are unlikely homes for living worlds. But recent simulations of the growth of such systems make it clear that a hot Jupiter isn’t necessarily a deal-breaker for habitable worlds. Are we going to have to add such systems into our target lists for future terrestrial planet finder missions?

Image: A hot Jupiter seared by its star. Could terrestrial planets survive the migration of such a planet inward from its place of origin in the outer system? Credit: Harvard-Smithsonian Center for Astrophysics.

Sean Raymond (University of Colorado), Avi Mandell and Steinn Sigurðsson (the latter two from the University of Pennsylvania) have been modeling circumstellar disks, beginning with a disk containing seventeen Earth masses of rocky/icy material divided between 80 ‘planetary embryos’ and 1200 planetesimals. The inner disk was modeled as iron-rich and water-poor, the outer as water-rich and iron-poor. These are values that are much like the Solar System, where it is believed that comets are as much as half water ice, and asteroids outside 2.5 AU contain large amounts of water.

Given these initial disk conditions, the team ran a Jupiter-mass planet from 5.2 AU in to 0.25 AU over a period of 100,000 years, studying the effects on planetary formation in the following 200 million years under a variety of scenarios. The first result is eye catching: A lot of things start to happen inside the orbit of the migrating giant planet, where planetesimals and embryos are drawn (‘shepherded’) by motion resonances and gas drag. ‘Hot Earths’ with masses up to five times that of Earth form relatively quickly.

Then something equally interesting occurs. Materials that have been scattered outward by the giant planet’s migration begin to have their orbits re-circularized by gas drag from the disk. Planetesimals outside the orbit of the ‘hot Jupiter’ begin to deliver materials including large amounts of water to growing terrestrial planets. Note this (internal references omitted for brevity):

Planets formed in the Habitable Zone in a significant fraction of our simulations. These planets have masses and orbits similar to those seen in previous simulations including only outer giant planets… However, Habitable Zone planets in systems with close-in giant planets tend to accrete a much larger amount of water than those in systems with only outer giant planets… The reason for these high water contents is twofold: 1) strong radial mixing is induced by the giant planet’s migration, and 2) in-spiraling icy planetesimals are easily accreted by planets in the terrestrial zone. Although we have not taken water depletion into account, these planets contain about twenty times as much water as those formed in similar outer giant planet simulations… These planets are likely to be “water worlds”, covered in kilometers-deep global oceans…

Interesting, no? We see the possibility of Earth-like planets (along with ‘hot Earths’ inside the gas giant’s orbit) that are water-rich, in some cases containing more than 100 times the water content of Earth, and low in iron (the iron content being diluted by outer, water-rich material). The influx of ices via outer planetesimals is unimpeded by the kind of scattering that Jupiter provides in our Solar System.

Out of all this comes a rough limit on the orbital distance of the inner giant planet that would allow the formation of terrestrial planets in the habitable zone: A terrestrial planet just inside the outer edge of the zone at 1.5 AU would demand the hot Jupiter’s orbit to be inside about 0.5 AU, although up to 0.7 AU may still be feasible. And when the researchers applied this limit to the known extrasolar giant planets, they found that 65 out of 178 of the systems in their sample permitted an Earth-like planet of at least 0.3 Earth masses or more to form in the habitable zone.

In other words, some of the known ‘hot Jupiter’ systems may turn out to yield potentially habitable planets after all, a key thought when drawing up target lists for next-generation planet-finder missions. “Our results suggest that terrestrial planets can coexist with both close-in giant planets and giant planets in outer orbits,” write the authors, “expanding the range of planetary systems that should be searched with these upcoming missions.” The paper is Mandell, Raymond and Sigurðsson, “Formation of Earth-like Planets During and After Giant Planet Migration,” Astrophysical Journal Vol. 660, Issue 1 (May, 2007), pp. 823-844 (abstract). Citation amended from the original as per the comment below.

Comments on this entry are closed.

  • ljk April 21, 2008, 15:16

    A peculiarity of metal-poor stars with planets ?

    Authors: Misha Haywood

    (Submitted on 18 Apr 2008)

    Abstract: Stars with planets at intermediate metallicities ([-0.7,-0.2] dex) exhibit properties that differ from the general field stars. Thirteen stars with planets reported in this metallicity range belong to the thick disc, while only one planet have been detected among stars of the thin disc. Although this statistics is weak, it contradicts the known correlation between the presence of planet and metallicity.

    We relate this finding to the specific property of the thin disc in this metallicity range, where stars are shown to rotate around the Galaxy faster than the Sun. Their orbital parameters are conveniently explained if they are contaminants coming from the outer Galactic disc, as a result of radial mixing. This must be considered together with the fact that metal-rich stars ([Fe/H]>+0.1 dex) found in the solar neighbourhood, which are the hosts of most of the detected planets, are suspected of being wanderers from the inner Galactic disc.

    It is then questionned why stars that originate in the inner and outer thin disc show respectively the highest and lowest rate of detected planets. It is suggested that the presence of giant planets might be primarily a function of a parameter linked to galactocentric radius, but not metallicity. Combined with the existing radial metallicity gradient, then radial mixing explains the correlation at high metallicity observed locally, but also the peculiarity found at low metallicity, which cannot be accounted for by a simple correlation between metallicity and planet probability.

    Comments: 4 pages, 2 figures, accepted for publication in A&A

    Subjects: Astrophysics (astro-ph)

    Cite as: arXiv:0804.2954v1 [astro-ph]

    Submission history

    From: Misha Haywood [view email]

    [v1] Fri, 18 Apr 2008 07:25:43 GMT (72kb)


  • andy April 21, 2008, 16:01

    Isn’t this old news?

    This paper you’ve linked is Mandell, Raymond and Sigurðsson “Formation of Earth-like Planets During and After Giant Planet Migration”, which was published in The Astrophysical Journal, Volume 660, Issue 1, pp. 823-844 (05/2007).

    Raymond, Mandell and Sigurðsson “Exotic Earths: Forming Habitable Worlds with Giant Planet Migration” (arXiv) is a different paper, published in Science, Vol. 313. no. 5792, pp. 1413 – 1416 (09/2006).

  • Adam April 21, 2008, 16:42

    Hi All

    According to the statistical analysis of Debra Fischer and colleagues the Hot Jupiters become more prevalent with super-Solar metallicity – it’s a very strong probability peak. Seems as the Galaxy ages – assuming age and metallicity are correlated that is – the terrestrials that form get “wetter” as Hot Jovians become more prevalent.

    There seem to be some interesting chemical evolution effects shaping the prospects for planets in our Galaxy. That same paper by Fischer et al also indicated low mass planets around most stars – BUT the planet assembly simulations of Sean Raymond et al imply that many such systems will result in DRY planets i.e. essentially zero initial water. Venus is potentially one such world, so I do wonder how lucky we were to have metallicity just right for a well-behaved Jupiter. Too low, and Earth would be like Venus. Too high and Earth would be Ocean, with an occasional volcano poking through.

  • Administrator April 21, 2008, 17:40

    Right you are, andy. I had been working with both papers open on the desktop and slipped in the wrong title. I also amended the citation.

  • James M. Essig April 21, 2008, 20:12

    Hi Folks;

    It would be interesting if gas giants could form say within 0.1 AU of a star with 1 percent the absolute luminosity of the Sun. Such gas giants might have moons capable of supporting life.

    Alternatively, if we can find gas giants at 1 AU within a Sun class star, a G-2 spectral type, then perhaps much closer scruitiny of the system for moons and for life, perhaps even intellegent life, on the moons would be in good order.

    Detecting gas giants in habitable zones arond G class stars, or perhaps even around red dwarfs might be a very good indicator of where to start looking for signs of life, perhaps even intellegent life.



  • philw1776 April 21, 2008, 20:47

    Somewhere across the galaxy on a beautiful planet briming with life and a nascent technical civilization, astronomers are speculating about the likelihood of other life bearing planets. Such alien worlds must of course have the requisite Hot Jovian which was instrumental in their solar system’s very early evolution.

  • Ron S April 21, 2008, 21:55

    If Earth were perfectly smooth (no continents or ocean basins), the volume of the oceans and fresh water would cover the planet with water to a depth of about 2,750 meters.

    So… does this make Earth a water planet, or should there be more even more water? It’s tectonic activity that allows for bits of rock to poke up here and there – currently 29% of the surface, and probably much lower during some earlier periods.

  • kurt9 April 22, 2008, 0:19

    As you may or may not know, the Central African rain forest is 50-60 million years old. This means that it has had plenty of time for highly evolved microorganism and parasites to emerge and render such areas uninhabitable to humans. There are certain areas of Central Africa where there are no people because these areas have highly evolved parasites and diseases that kill people who try to live in these areas. The Amazon in South America is much newer, in geological age and, thus, does not contain uninhabitable areas like Central Africa.

    This suggests that a certain percentage of the “Earth-like” worlds with highly evolved lifeforms on them may, indeed, be quite toxic and poisonous to human travelers who land on them for a visit (assuming we develop some form of FTL to get to them). There may even be planets out there that look like they are habitable when viewed from afar, but have an ecosystem similar to that of the Chtorr. Such worlds would certainly be uninhabitable to humans even though they have the right temperature and breathable atmosphere.

    Speaking of hot Jupiters, why didn’t ours migrate in like the ones we seem to be finding these days?

  • Ronald April 22, 2008, 2:30

    Further to Adam; assuming that ‘just right’ does not exist in planet formation, I wonder what the right metallicity *range* is: not so low that it results in a failed planetary system or dry terrestrial planets, not so high that it results in only water worlds or worse: only giant planets. I guess somewhere between (0.3?) 0.5 – 1.5 (2?) times solar. We’ll find out some time in the future.

    Funny in a way: some day metallicity may appear to be a more important factor in ‘the right’ terrestrial planet formation than the ‘right’ stellar spectral type. Though the fact that terrestrial planets may still form with inward migrated hot Jupiters is quite encouraging.

  • Administrator April 22, 2008, 7:32

    kurt9, re our own gas giants and why we have no ‘hot Jupiter’ here, I’ll be writing about this question later today.

  • Richard April 22, 2008, 16:33

    Kurt9 says, “highly evolved parasites and diseases that kill people”.

    My understanding of current biological thinking is that highly evolved parasites and diseases are relatively benign, because there is no particular advantage in a parasite killing its home and food source.

    Parasites and viruses on an alien world might be highly evolved for their hosts, but they might not even be able to infect human bodies because they haven’t evolved to recognize humans as hosts.

    I think a bigger problem might be that alien worlds might have a biochemistry which would be so incompatible with ours that we would find nothing to eat there. They might use amino acids that earth life doesn’t use. We would find ourselves trying to grow crops on a planet full of noxious and inedible weeds which are better adopted to local conditions than our crop plants.

  • Ronald April 23, 2008, 2:06

    @Kurt9, off-topic: is the Central African rainforest really this old? I actually think that most of it is much younger, only having spread after the last ice age, during which most of Central Africa was much drier and rainforest was reduced to certain refuges (refugia).
    But having said that, then of course it is possible that part of it, i.e. the species assemblages and ecosystems per se, can be much older.
    But older than Amazonia? The Amazon rainforest is much species richer than
    @Central Africa, and so is the Indo-Malaysian rainforest.

    @philw1776: I really like your posting, because it nicely describes our human geo-centrism. What we know and are simply used to should always be distinguished from the objective physical requirements, restrictions and possibilities with regard to planetary life.

  • Ronald April 23, 2008, 8:22

    @Kurt9: on second thought, I think it is even the other way around, concerning the Amzon and Central Africa: the Americas evolved until very recently without humans, hence there were (until Europenas brought them) relatively few diseases and parasites that could affect humans, because almost none co-evolved.

    In Africa it was the opposite: because of long human presence, so many diseases and parasites co-evolved. It is long human presence that made Central Africa so inhospitable, rather than the opposite.

    An alien planet would probably pose less danger to us humans, at least as far as diseases are concerned (big critters might be a different thing), than Dutch Elm’s disease, because the local micro-organisms would not recognize us as food.

  • James M. Essig April 24, 2008, 21:17

    Hi Ronald;

    Interesting discussion.

    Speaking of big critters, it occurred to me that alien animals might pose some danger to human crew members landing on any planets with animal life especially lifeforms that are not DNA based and have flesh or bodies made of tissue that is as strong as Kevlar or the material Zylon which has about 2.5 times the tensile strength of Kevlar. Alternatively, very large ET animals might have body tissues with the strength of carbon nanotubes and made of diamond like carbon. Diamond has an extremely high energy of vaporization relative to its solid state at room temperature.

    Dealing with attacks from such animals might require quick reaction times on the part of the ground crews with shaped charged explosive rounds of the same or greater power than those that can hold the modern U.S. M1-A2 Ambrams battle tanks at risk. I think we could deal with such animals, but man, I would hate to be surprised by a huge snake made out of such super strong material.

    The preferable solution, however, would be not to upset or provoke the animals by showing such animals gentleness or effectionate guestures much as animal experts such as Jane Goodall manage to do with the large and very powerful gorillas that she lived among in order to study them more closely. It would be very wise to include animal psychologists as landing crew members including theoretical animal psychologists.



  • ljk April 30, 2008, 11:51

    HD 75289Ab revisited – Searching for starlight reflected from a hot Jupiter

    Authors: F. Rodler, M. Kuerster, T. Henning

    (Submitted on 29 Apr 2008)

    Abstract: Aims. We attempt to detect starlight reflected from a hot Jupiter, orbiting the main-sequence star HD 75289Ab. We report a revised analysis of observations of this planetary system presented previously by another research group.

    Methods. We analyse high-precision, high-resolution spectra, collected over four nights using UVES at the VLT/UT2, by way of data synthesis. We try to interpret our data using different atmospheric models for hot Jupiters.

    Results. We do not find any evidence for reflected light, and, therefore, establish revised upper limits to the planet-to-star flux ratio at the 99.9% significance level. At high orbital inclinations, where the best sensitivity is attained, we can limit the relative reflected radiation to be less than e = 6.7 x 10-5 assuming a grey albedo, and e = 8.3 x 10-5 assuming an Class IV function, respectively. This implies a geometric albedo smaller than p = 0.46 and p = 0.57, for the grey albedo and the Class IV albedo shape, respectively, assuming a planetary radius of 1.2 RJup.

    Comments: accepted by A&A

    Subjects: Astrophysics (astro-ph)

    Cite as: arXiv:0804.4609v1 [astro-ph]

    Submission history

    From: Florian Rodler [view email]

    [v1] Tue, 29 Apr 2008 13:48:26 GMT (28kb)


  • ljk May 13, 2008, 14:57

    Turbulence in Extrasolar Planetary Systems Implies that Mean Motion Resonances are Rare

    Authors: Fred C. Adams, Gregory Laughlin, Anthony M. Bloch

    (Submitted on 12 May 2008)

    Abstract: This paper considers the effects of turbulence on mean motion resonances in extrasolar planetary systems and predicts that systems rarely survive in a resonant configuration. A growing number of systems are reported to be in resonance, which is thought to arise from the planet migration process. If planets are brought together and moved inward through torques produced by circumstellar disks, then disk turbulence can act to prevent planets from staying in a resonant configuration.

    This paper studies this process through numerical simulations and via analytic model equations, where both approaches include stochastic forcing terms due to turbulence. We explore how the amplitude and forcing time intervals of the turbulence affect the maintenance of mean motion resonances. If turbulence is common in circumstellar disks during the epoch of planet migration, with the amplitudes indicated by current MHD simulations, then planetary systems that remain deep in mean motion resonance are predicted to be rare.

    More specifically, the fraction of resonant systems that survive over a typical disk lifetime of 1 Myr is of order 0.01. If mean motion resonances are found to be common, their existence would place tight constraints on the amplitude and duty cycle of turbulent fluctuations in circumstellar disks. These results can be combined by expressing the expected fraction of surviving resonant systems in the approximate form P_b = C / N_{orb}^{1/2}, where the dimensionless parameter C = 10 – 50 and where N_{orb} is the number of orbits for which turbulence is active.

    Comments: 30 pages, 5 figures, accepted to ApJ

    Subjects: Astrophysics (astro-ph)

    Cite as: arXiv:0805.1681v1 [astro-ph]

    Submission history

    From: Fred C. Adams [view email]

    [v1] Mon, 12 May 2008 16:40:14 GMT (38kb)


  • ljk June 3, 2008, 14:47

    Predicting low-frequency radio fluxes of known extrasolar planets

    Authors: J.-M. Grießmeier (1), P. Zarka (1), H. Spreeuw (2) ((1) LESIA, Observatoire de Paris, Meudon, France, (2) Astronomical Institute ”Anton Pannekoek”, Amsterdam, The Netherlands)

    (Submitted on 2 Jun 2008)

    Abstract: Context. Close-in giant extrasolar planets (”Hot Jupiters”) are believed to be strong emitters in the decametric radio range.

    Aims. We present the expected characteristics of the low-frequency magnetospheric radio emission of all currently known extrasolar planets, including the maximum emission frequency and the expected radio flux. We also discuss the escape of exoplanetary radio emission from the vicinity of its source, which imposes additional constraints on detectability.

    Methods. We compare the different predictions obtained with all four existing analytical models for all currently known exoplanets. We also take care to use realistic values for all input parameters.

    Results. The four different models for planetary radio emission lead to very different results. The largest fluxes are found for the magnetic energy model, followed by the CME model and the kinetic energy model (for which our results are found to be much less optimistic than those of previous studies). The unipolar interaction model does not predict any observable emission for the present exoplanet census. We also give estimates for the planetary magnetic dipole moment of all currently known extrasolar planets, which will be useful for other studies.

    Conclusions. Our results show that observations of exoplanetary radio emission are feasible, but that the number of promising targets is not very high. The catalog of targets will be particularly useful for current and future radio observation campaigns (e.g. with the VLA, GMRT, UTR-2 and with LOFAR).

    Comments: 4 figures; Table 1 is available in electronic form at the CDS via anonymous ftp to cdsarc.u-strasbg.fr ( or via this http URL

    Subjects: Astrophysics (astro-ph)

    Journal reference: A&A 475, 359-368 (2007)

    DOI: 10.1051/0004-6361:20077397

    Cite as: arXiv:0806.0327v1 [astro-ph]

    Submission history

    From: Jean-Mathias Grie{\ss}meier [view email]

    [v1] Mon, 2 Jun 2008 17:26:55 GMT (73kb)


  • ljk June 11, 2008, 23:17

    Assembling the Building Blocks of Giant Planets around Intermediate Mass Stars

    Authors: K. A. Kretke, D. N. C. Lin, P. Garaud, N. J. Turner

    (Submitted on 9 Jun 2008)

    Abstract: We examine a physical process that leads to the efficient formation of gas giant planets around intermediate mass stars. In the gaseous protoplanetary disks surrounding rapidly-accreting intermediate-mass stars we show that the midplane temperature (heated primarily by turbulent dissipation) can reach > 1000 K out to 1 AU.

    Thermal ionization of this hot gas couples the disk to the magnetic field, allowing the magneto-rotational instability (MRI) to generate turbulence and transport angular momentum. Further from the central star the ionization fraction decreases, decoupling the disk from the magnetic field and reducing the efficiency of angular momentum transport.

    As the disk evolves towards a quasi-steady state, a local maximum in the surface density and in the midplane pressure both develop at the inner edge of the MRI-dead zone, trapping inwardly migrating solid bodies. Small particles accumulate and coagulate into planetesimals which grow rapidly until they reach isolation mass.

    In contrast to the situation around solar type stars, we show that the isolation mass for cores at this critical radius around the more massive stars is large enough to promote the accretion of significant amounts of gas prior to disk depletion. Through this process, we anticipate a prolific production of gas giants at ~1 AU around intermediate-mass stars.

    Comments: 24 pages with 5 figures, preprint format. Submitted to ApJ

    Subjects: Astrophysics (astro-ph)

    Cite as: arXiv:0806.1521v1 [astro-ph]

    Submission history

    From: Katherine Kretke [view email]

    [v1] Mon, 9 Jun 2008 19:55:14 GMT (67kb)


  • ljk June 13, 2008, 15:15

    Evolution of Migrating Planets Undergoing Gas Accretion

    Authors: Gennaro D’Angelo, Stephen H. Lubow

    (Submitted on 11 Jun 2008)

    Abstract: We analyze the orbital and mass evolution of planets that undergo run-away gas accretion by means of 2D and 3D hydrodynamic simulations. The disk torque distribution per unit disk mass as a function of radius provides an important diagnostic for the nature of the disk-planet interactions.

    We first consider torque distributions for nonmigrating planets of fixed mass and show that there is general agreement with the expectations of resonance theory. We then present results of simulations for mass-gaining, migrating planets. For planets with an initial mass of 5 Earth masses, which are embedded in disks with standard parameters and which undergo run-away gas accretion to one Jupiter mass (Mjup), the torque distributions per unit disk mass are largely unaffected by migration and accretion for a given planet mass.

    The migration rates for these planets are in agreement with the predictions of the standard theory for planet migration (Type I and Type II migration). The planet mass growth occurs through gas capture within the planet’s Bondi radius at lower planet masses, the Hill radius at intermediate planet masses, and through reduced accretion at higher planet masses due to gap formation. During run-away mass growth, a planet migrates inwards by only about 20% in radius before achieving a mass of ~1 Mjup.

    For the above models, we find no evidence of fast migration driven by coorbital torques, known as Type III migration. We do find evidence of Type III migration for a fixed mass planet of Saturn’s mass that is immersed in a cold and massive disk. In this case the planet migration is assumed to begin before gap formation completes. The migration is understood through a model in which the torque is due to an asymmetry in density between trapped gas on the leading side of the planet and ambient gas on the trailing side of the planet.

    Comments: 26 pages, 29 figures. To appear in The Astrophysical Journal vol.684 (September 20, 2008 issue)

    Subjects: Astrophysics (astro-ph)

    DOI: 10.1086/590904

    Cite as: arXiv:0806.1771v1 [astro-ph]

    Submission history

    From: Gennaro D’Angelo Dr. [view email]

    [v1] Wed, 11 Jun 2008 00:38:38 GMT (445kb)


  • ljk July 30, 2008, 19:49

    Habitable Climates: The Influence of Obliquity

    Authors: David S. Spiegel, Kristen Menou, Caleb A. Scharf

    (Submitted on 25 Jul 2008)

    Abstract: Without the stabilizing influence of the Moon, the Earth’s obliquity could vary significantly. Extrasolar terrestrial planets with the potential to host life may therefore have large obliquities or be subject to strong obliquity variations.

    We revisit the habitability of oblique planets with an energy balance climate model (EBM) allowing for dynamical transitions to ice-covered snowball states as a result of ice-albedo feedback. Despite the great simplicity of our EBM, it captures reasonably well the seasonal cycle of global energetic fluxes at Earth’s surface. It also performs satisfactorily against a full-physics climate model of a highly oblique Earth, in an unusual regime of circulation dominated by heat transport from the poles to the equator.

    Climates on oblique terrestrial planets can violate global radiative balance through much of their seasonal cycle, which limits the usefulness of simple radiative equilibrium arguments. High obliquity planets have severe climates, with large amplitude seasonal variations, but they are not necessarily more prone to global snowball transitions than low obliquity planets.

    We find that terrestrial planets with massive CO2 atmospheres, typically expected in the outer regions of habitable zones, can also be subject to such dynamical snowball transitions. Some of the snowball climates investigated for CO2-rich atmospheres experience partial atmospheric collapse. Since long-term CO2 atmospheric build-up acts as a climatic thermostat for habitable planets, partial CO2 collapse could limit the habitability of such planets. A terrestrial planet’s habitability may thus depend sensitively on its short-term climatic stability.

    Comments: 34 pages, 13 figures, submitted to ApJ

    Subjects: Astrophysics (astro-ph); Atmospheric and Oceanic Physics (physics.ao-ph)

    Cite as: arXiv:0807.4180v1 [astro-ph]

    Submission history

    From: David Spiegel [view email]

    [v1] Fri, 25 Jul 2008 20:18:50 GMT (518kb,D)