≡ Menu

Disruptive Planets and their Consequences

One of the joys of writing a site like Centauri Dreams is that I can choose my own topics and devote as much or as little time as I want to each. The downside is that when I’m covering something in greater depth, as with the four articles on antimatter that ran in the last six days, I invariably fall behind on other interesting work. That means a couple of days of catch-up, which is what we’ll now see, starting with some thoughts on a possible planet beyond Neptune, a full-sized world as opposed to an ice dwarf like Pluto or Eris. This story is actually making the rounds right now, but it triggered thoughts on older exoplanet work I’ll describe in a minute.

It’s inevitable that we call such a world Planet X, in my case because of my love for the wonderful Edgar Ulmer film The Man from Planet X (1951), in which a planet from the deeps wanders into the Solar System and all manner of trouble — including the landing of an extraterrestrial on a foggy Scottish moor — breaks out. Of course, Planet X was also the name for the world Percival Lowell was searching for in the 1920s, a hunt that resulted in Clyde Tombaugh’s discovery of Pluto, although the latter occurred more or less by chance since Pluto/Charon isn’t big enough to cause the gravitational effects Lowell was examining.

So is there a real Planet X? Rodney Gomes (National Observatory of Brazil) has run simulations on the ‘scattered disk’ beyond Neptune and, factoring in oddities like the highly elliptical orbit of Sedna and other data points on these distant objects, Gomes believes a Neptune-class planet about four times as massive as Earth may be lurking in the outer system. Sedna, you may recall, has a perihelion of 76 AU but an aphelion fully 975 AU out — it’s on a 12,000 year orbital period! As for Gomes, his team has been looking at what they call ‘true inner Oort cloud objects’ for some time, seeing objects like Sedna as markers for the existence of a planet.

Gomes ran through the results of his simulations at an American Astronomical Society meeting in Oregon in May, keeping the Planet X hunt alive, and it’s worth noting that a Jupiter-class planet at about 5000 AU may also fit the bill (see Finding the Real Planet X). For that matter, the orbits of scattered disk objects may have another explanation besides an undiscovered planet. But thinking about Gomes’ work brought me around to Jason Steffen and team, whose new paper goes to work on a much different kind of gravitational effect, the disruption caused by a ‘hot Jupiter’ as it moves through a young Solar System and scatters smaller planets.

Realm of the Wandering Planets

Steffen (Fermilab Center for Particle Astrophysics) is digging into exactly what makes ‘hot Jupiters’ take up such extreme orbits. These are planets of Jupiter’s size and larger that whip around their stars in periods of just a few days. The question is how they got to their present position, for the assumption is that planets of this size had to form much further out in their system and then move inward. There are two mechanisms that could make this happen, one of which — a slow migration through a gas disk that would allow low-mass planets to likewise migrate inward, where they can be captured into mean-motion resonance with the gas giant — seems benign. These models suggest the presence of smaller worlds near the hot Jupiter.

Image: Artist’s concept of a hot Jupiter, likely a disrupter of any planets that encounter it. Credit: NASA.

The other model is lethal to the inner system. Here, the giant planet’s migration is caused by gravitational interactions with another gas giant that result in one of the worlds being flung into interstellar space, while the other migrates inward and disrupts the orbits of any inner-system worlds. This scenario is what the Steffen paper is suggesting, for the team’s analysis of 63 Kepler planets around solar-type stars in orbits of 6.3 days or less shows no evidence at all for nearby planets. If such worlds were there, they ought to be detectable through transit timing variations (TTV) unless they are smaller than the Earth, or much further out in the system.

To compare and contrast environments, the researchers took another sample of 31 Kepler planets with ‘warm Jupiters’ — planets of Jupiter size around the same kind of star, but with longer orbital periods of between 6.3 and 15.8 days. They also checked 222 Kepler ‘hot Neptunes.’ The result: Three of the 31 ‘warm Jupiter’ systems showed companion planets in the inner system, and fully one-third of the hot Neptune systems showed the presence of inner system planets. Finally, the team looked at 52 ‘hot Earths’ in the Kepler data for TTVs, testing whether hot Jupiters and smaller worlds like these might co-exist in mutually inclined orbits. They found no evidence for high-mass companions on inclined orbits in this scenario.

The authors see this as a boost to the ‘scattering’ model, the study suggesting that hot Jupiters are migrating worlds on initially highly elliptical orbits that scattered other planets out of the inner system before their orbits became circularized close to their stars. Short period, low-mass planets would seem to have a different formation history than hot Jupiters. From the paper:

Hot Jupiter systems where planet-planet scattering is important are unlikely to form or maintain terrestrial planets interior to or within the habitable zone of their parent star. Thus, theories that predict the formation or existence of such planets (Raymond et al. 2006; Mandell et al. 2007) can only apply to a small fraction of systems. Future population studies of planet candidates, such as this, that are enabled by the Kepler mission will yield valuable refinements to planet formation theories — giving important insights into the range of probable contemporary planetary system architectures and the possible existence of habitable planets within them.

If hot Jupiter systems have a different dynamical history than other planetary systems, as this work suggests, then we have a useful filter to apply to exoplanet studies. If it can be firmly established that the presence of a hot Jupiter means no planets in the habitable zone, we know our resources are best focused elsewhere when it comes to looking for terrestrial worlds. It’s too early to make that call now, but the evidence is mounting that in most cases hot Jupiters are killer worlds when it comes to young planets in the warm regions where life may occur.

The paper is Steffen at al., “Kepler constraints on planets near hot Jupiters,” Proceedings of the National Academy of Sciences 109 (21) 7982-7987 (2012). Abstract available.


Comments on this entry are closed.

  • Dan May 24, 2012, 10:05

    Has the data from WISE placed any meaningful constraints on a Planet X yet?

  • Ronald May 24, 2012, 13:29

    Great post and summary of the Steffen’s paper.

    I would like to add to your sentence “If hot Jupiter systems have a different dynamical history than other planetary systems, as this work suggests, then we have a useful filter to apply to exoplanet studies”:

    And if hot Jupiter systems (and other types of planetary systems) can be characterized and diagnosed by their metallicity and specific abundance patterns, as other studies (Ramirez, Melendez, et al.) suggest, then we even have an indirect spectroscopic filter for exoplanet studies and the search for earthlike planets.

  • Dan Ibekwe May 24, 2012, 18:24

    Whatever became of the ‘Kuiper Cliff’, the idea that there seemed to be a sudden, sharp outer boundary to the Kuiper Belt at about 80 AU? (Presumably caused by the presence of an as yet undetected Neptune-sized body).

    Has it been superseded by more recent observations?

  • Joy May 24, 2012, 23:01

    Re Gomes: Well, non-specific predictions are becoming quite fashionable now, aren’t they. Gomes has allowed his error bars to stretch astronomical distances, from NatGeo:

    “Based on his calculations, Gomes thinks a Neptune-size world, about four times bigger than Earth, orbiting 140 billion miles (225 billion kilometers) away from the sun—about 1,500 times farther than Earth—would do the trick. But so would a Mars-size object—roughly half Earth’s size—in a highly elongated orbit that would occasionally bring the body sweeping to within 5 billion miles (8 billion kilometers) of the sun.
    Also, Gomes’s simulations don’t give astronomers any clue as to where to point their telescopes—”it can be anywhere,” he said.”

    Utterly useless work. The string theorists have shown the way, testable predictions are no longer relevant in postmodern thought, as long as ideas have some ability to draw eyeballs and click throughs on the interwebs.

    @Dan Ibekwe – The Kuiper Cliff is still there, however various explanations for it have been mooted, the presence of an as yet undetected Neptune-sized body being among the less plausible

  • Paul May 25, 2012, 9:00

    I would very much like there to be a smallish planet in the outer solar system. A body the size of Mars or larger may be able to retain helium against loss to space (depending on the temperature of the exobase of its atmosphere), which could make such a planet the easiest source of 3He in the solar system, assuming we have a fusion rocket that could get out to it in reasonable time. Being small, the planet would be easier to operate on than, say, Uranus.

  • A. A. Jackson May 25, 2012, 10:27

    The “Kuiper cliff” is still a mystery.
    Since the suggestion that a Planet X beyond 100 AU might be the cause, I don’t know if much work being done.
    The standard model is still that the Oort cloud is a product of the ‘last scattering’ by mostly Jupiter … also the outer gas giants. That cloud does have a ‘outer shell’. If you look at the Wiki picture the ‘inner cloud’ is quite complex.
    Yet we still don’t know if it’s there… between about 1000 AU to 10,000 AU we can’t see yet.

    The Kuiper Belt had a funny history , we submitted a paper in 1987 to a journal in 1987 about the infrared brightness of the Kuiper Belt, which was rejected because a referee said there was no Kuiper belt! We changed the name and submitted to a different journal. It got published,

    Infrared Brightness of a Comet Belt Beyond Neptune, A. A. Jackson and R.M.Killen, Earth, Moon and the Planets, vol.42, 41, 1988

    Tho Gerard Kuiper predicted it on the basis of solar nebula formation in 1951 (preceded by Kenneth Edgeworth) it was not confirmed until 1992.

  • ljk May 25, 2012, 14:05

    A. A. Jackson wrote on May 25, 2012 at 10:27:

    “The Kuiper Belt had a funny history , we submitted a paper in 1987 to a journal in 1987 about the infrared brightness of the Kuiper Belt, which was rejected because a referee said there was no Kuiper belt! We changed the name and submitted to a different journal. It got published!”

    Did you ever contact the referee after 1992 and say “Do you like us now?!”

    The September, 1992 issue of Astronomy magazine had a story asking if the Kuiper Belt existed or not. And the very next month the first one was discovered.

  • Eniac May 25, 2012, 23:29

    While this seems unpopular here, I would like to defend string theory. The theory is not untestable, because its whole reason for being is to explain things that we can not yet explain, such as the masses of the elementary particles, or quantum gravity, or dark matter and dark energy, or anything else I forgot about. If it accurately and elegantly predicts any of these things that have already been experimentally observed, that will be a test passed, and a great success. I have no idea if string theory in particular is likely to achieve this, but I do not think the negative comments around here are any more informed than that, and no better alternative has been offered.

  • Interstellar Bill May 26, 2012, 2:00

    Very soon after the first hot Jupiter was found at 51 Pegasi, speculation began that their migration would get rid of smaller planets they encountered. Many simulations didn’t show this, but thanks to Kepler we now have some data, with more on the way.

    Kepler’s findings are as exciting as our first closeups of outer-planet moons, or the Martian surface from a lander. How about all those regularly spaced planets crammed inside a Venus-size orbit? Long live Kepler!

  • andy May 27, 2012, 15:06

    Way to trivialise the result for some cheap mockery, Joy. But mocking scientists is becoming quite fashionable n0w isn’t it? I presume the studies that predicted a planet in the Beta Pictoris system from dust ring morphology were utterly useless work as well, as they couldn’t predict the precise point on its orbit that the hypothetical planet would have. Obviously with Beta Pictoris we have the advantage that we are not sitting inside the system with the whole sky to look at, but the principle is similar. Exploring the allowable parameter space may not lead directly to the glory of a planet detection, but that does not make it useless.

  • ljk May 28, 2012, 10:31

    I have read that to detect a cosmic string, it would take a particle accelerator the circumference of the Milky Way galaxy to do the job. Hopefully we will find slightly smaller and less expensive methods to detect strings, but I also have to agree that it is a problem for science to be unable to detect or test for their existence experimentally. Math is not proof (pun intended).

  • cshin9 August 26, 2012, 1:24

    Have there been any more discoveries of objects in Sedna-like orbits? And if so, is there any tendency towards one direction of the sky?