≡ Menu

A Comparative Look at Solar Systems

With almost 2000 exoplanets now confirmed, not to mention candidates in the thousands, it’s amazing to recall that it was just twenty years ago that the first planet orbiting a main sequence star beyond the Solar System was found. Continued work on the world revealed that 51 Pegasi b is about half as massive as Jupiter, though 50 percent larger. Orbiting its star in roughly four days, the planet is some fifty light years from Earth. Thus we began to learn not just that exoplanets were out there, but that their environments could be truly extreme — remember that it was just in 1992 that planets were found around the pulsar PSR 1257+12.

Without any evidence other than my imagination, I grew up believing that most stars should have planets, and just assumed that their stellar systems were more or less like our own. There should be a few planets too close to their star for life to exist, and several gas giants out at the outskirts of the system, and somewhere in between there should be a world not so different from Earth. It was a naive view, but not completely implausible, and anyway, we lacked data.

Discovering ‘hot Jupiters’ is one way we began to realize that other configurations could exist, and the number of ‘super-Earths’ has made the same case. Just how ‘normal’ is our Solar System in the first place? A new paper from Rebecca Martin (University of Nevada, Las Vegas) and Mario Livio (Space Telescope Science Institute) tackles the question, comparing what we see in our Solar System to our growing database of exoplanetary information.

Obviously, this is a work in progress, for we’re not only still examining abundant Kepler and K2 data but continuing a robust planet hunt that looks forward to space-based missions like TESS (Transiting Exoplanet Survey Satellite) and PLATO (PLAnetary Transits and Oscillations of stars), surveys of considerable scope that should build our catalog of ‘nearby’ planets. Nonetheless, we can draw some conclusions based on what we see in the image below.


Image (click to enlarge): Model of all the multi-planet systems found by Kepler as of November 2013; our terrestrial planets are shown in grey at the top left for comparison. A new study examines how our solar system compares to the exoplanetary systems we’ve found. Credit: NASA/Kepler/Dan Fabricky.

Our Solar System is composed of a good deal more than planets, of course, which leads to an important caveat, one that Martin and Livio mention early on in the paper. We have two belts orbiting the Sun, the main asteroid belt and the Kuiper Belt. The problem is that although we can see a number of debris and dust belts around other stars, belts with as little mass as ours would not be observable to us around other stars. So a study like this one has to base its findings solely on planets. It can be mentioned, though, that hundreds of debris disk candidates are now in play, and about two-thirds of these are best modeled as two component disks.

That’s a plus for the idea that our Solar System isn’t all that atypical. What about the planets? The authors use a mathematical transformation that allows them to set up a statistical comparison. The low mean eccentricity of planets around our Sun is one area where we differ from other multi-planet systems — our planets move in largely circular orbits — but as the paper notes, our observation methods are biased toward finding high eccentricity planets. Circular orbits work to our benefit, for planets with low eccentricity are more likely to be dynamically stable. Indeed, the terrestrial planets in our system are thought to be stable until that distant time when the Sun becomes a red giant and disrupts the entire inner system.

What about age? The Sun is about half the age of the Milky Way disk, hardly setting up our system as special, and at least one study has found that about 80 percent of existing Earth-like planets were already formed when the Earth came into existence. We also know that terrestrial planets in the habitable zone of their host star appear to be common. The paper notes, for example, the work of Courtney Dressing and David Charbonneau, which uses Kepler data for M-dwarfs and finds an occurrence rate for Earth-sized planets in the habitable zone of 18% to 27%, a conservative estimate that Martin and Livio say could be as high as 50 percent (see How Common Are Potential Habitable Worlds in Our Galaxy?)

If we’re not unusual in terms of age or habitability, we do differ considerably from other systems in two respects. First, we have no planets inside the orbit of Mercury, in contrast to systems with rocky worlds on far closer orbits. Moreover, the Solar System lacks a super-Earth, a category of planet now turning out to be common.

The paper summarizes its findings this way:

We find that the properties of the planets in our solar system are not so significantly special compared to those in exosolar systems to make the solar system extremely rare. The masses and densities are typical, although the lack of a super-Earth-sized planet appears to be somewhat unusual. The orbital locations of our planets seem to be somewhat special but this is most likely due to selection effects and the difficulty in finding planets with a small mass or large orbital period. The mean semi-major axis of observed exoplanets is smaller than the distance of Mercury to the Sun. The relative depletion in mass of the solar system’s terrestrial region may be important. The eccentricities are relatively low compared to observed exoplanets, although the observations are biased toward finding high eccentricity planets. The low eccentricity, however, may be expected for multi-planet systems. Thus, the two characteristics of the solar system that we find to be most special are the lack of super-Earths with orbital periods of days to months and the general lack of planets inside of the orbital radius of Mercury.

So while we’ve had quite a few surprises in the past twenty-five years, going from no exoplanets known to planets around pulsars and then main sequence stars, and moving from those early detections to thousands of candidates, we’re not seeing anything that would peg us as being unique. In terms of habitability, the authors see nothing in the Solar System that would make it especially conducive to life’s formation as opposed to other planetary systems:

If exosolar life happens to be rare it would probably not be because of simple basic physical parameters, but because of more subtle processes that are related to the emergence and evolution of life. Since at the moment we do not know what those might be, we can allow ourselves to be optimistic about the prospects of detecting exosolar life.

That lack of a super-Earth troubles me, though. Systems that have a super-Earth generally have more than one. The authors ask a good question: Does the presence of a super-Earth affect terrestrial planet formation? Several studies have looked at a migrating super-Earth moving slowly through the habitable zone, finding that a terrestrial planet that forms there later will tend to be rich in volatiles. Many observed super-Earths are found in orbits where they were unlikely to have formed, so scenarios of super-Earth migration surely deserve further study.

The paper is Martin and Livio, “The Solar System as an Exoplanetary System,” The Astrophysical Journal Vol. 810, No. 2 (3 September 2015). Full text.


Comments on this entry are closed.

  • Brett October 8, 2015, 15:45

    I’d be more comfortable once we manage to regularly detect earth-sized (or smaller) planets around stars. If the frequency of super-earths in systems is still extremely high, then we could say the solar system is unusual. If it’s not, well . . .

  • RobFlores October 8, 2015, 15:58

    If I may point out one more extra-solar inferences if not outright findings.

    Multi-planet Exo solar systems demonstrate orbits more compact than what was expected when the search for planets began. I am amazed that 3-4 fairly large planets can dance inside of a Venus orbit. I never thought such
    arrangements would survive a migration or even form close to their primary.

    So far the data can also confirm that Saturn-Jupiter sized Jovians are not dominant in orbits close to their primary Considering the find at 51 Pegasi this was certainly a relief to those looking for twin earths.

    Regarding the Lack of Super-terrestrial in our solar system. I have not come across any inference or postulate that shows a natural skewing towards
    the formation of ST and Mini Neptunes. The most plausible reason for our
    lack of large terrestrial maybe due to chance. My guess is that when looked at part of larger sample with a better future survey study we will find that planets Mars sized to Mini-Neptunes exist in a continuum and are distributed fairly evenly on the whole.

  • Tom Mazanec October 8, 2015, 18:27

    Actually I understand there is about a 1% chance Mercury will collide with Venus because of Jupiter increasing it’s eccentricity in a few billion years.

  • P October 8, 2015, 21:35

    I’m with Brett, we need to get our RV sensitivities up, and continue to track known planetary systems for many more years to see if true Earth sized planets rise above the noise floor. I know there are ongoing arguments about this, many of which you’ve discussed here. What if many systems where currently only giants are known are eventually found to have ‘Earths’? What if systems with known Super Earths are eventually found to have Earths as well? Many systems have large ‘gaps’ where many small and currently undetectable planets could fit. Transit surveys probably aren’t going to reveal many of long period ‘Earths’ around sunlike stars – the transit probability is just too low..

    We need more time and more data. I’m looking for TESS to make a contribution regarding ‘Earths’ around nearby stars, albeit with mostly M and K stars.


  • James Stilwell October 9, 2015, 10:17

    The earth-moon may turn out to be a rare system and of significant importance in the development of life on earth…The moon will become an indispensible stepping stone in large scale human movement into the cosmos…No mention here of the importance of the moon is interesting…Discussing why a large moon would be of little or no importance to life evolving on earth would also be very interesting…I want to learn more…Discussions of earth twins would need to include the presence of a 2000 mile diameter moon or it is not exactly an earth twin…Hasn’t the presence of our large moon moderated the spin of the earth on its axis?
    Could plant life have evolved on an earth spinning at ten times it’s present speed?

  • R Kelley October 9, 2015, 10:54

    The article “Terrestrial Planet Formation in the Presence of Migrating Super-Earths” (Izidoro, Morbidelli, Raymond) indicates that:
    1. The presence of migrating Super-Earths in close orbit around stars is most likely explained by migration instead of in-situ formation
    2. Provided such Super-Earths pass “quickly” through the habitable zone, terrestrial planets will likely still form after such passage. This is true even if you are talking about multiple Super-Earths passing through the habitable zone.
    3. “Quickly” for this purpose seems to be a passage from the original location of formation (beyond the snow-line) to a close solar orbit in less than one million years.

    I am not aware of studies which have determined or estimated how quickly Super-Earths migrate. It may be that the passage of Neptune size or larger bodies would wreck havoc on materials that could form terrestrial planets regardless of how quick the passage is through the habitable zone.

    Since 50% or more of observed solar systems have Super-Earths in close orbit, such objects clearly should be considered in determining the probability that earth-like bodies exist in the habitable zone.

  • Alex Tolley October 9, 2015, 12:09

    @James Stillwell – bear in mind that life was prokaryotic for most of the life era. I am not aware that bacteria, especially those living in sunless areas are impacted by day-night cycles, or spin rate. So I don’t consider a moon as essential for the genesis of life.

    Certainly Earth life has evolved with the moon, and some life forms are dependent on the moon, most famously the grunion. Many plants do require a day-night cycle, although many will be OK under continuous light conditions. Others require seasonal cycles. If the Earth was less stable on its spin axis, then this signal wouldn’t be suitable for evolution to act upon.

    My bottom line is that I don’t see the Moon as necessary for life to evolve, even multi-cellular life. The absence of such a body will impact the conditions under which life evolves.

  • Stan Erickson October 9, 2015, 13:24

    @James Stilwell: your question about the effects of a large moon on life origination is very apt, and the answer is likely to be very interesting. One effect of the moon comes from the larger angular momentum it has, making it hard to tilt its orbit. This then has a pull on the axial rotation of the Earth, keeping it much more stable than otherwise. So, could life evolve on a planet where the axial tilt varied a large amount? One thing I find most un-understandable: does the moon have an effect on the magnetic field of the Earth? The connection between the magnetosphere and life origination is fairly clear, so this is a route for a lunar effect on life.

  • I Brouwer October 10, 2015, 10:28

    the moon earth relation is intriguing, as an earth without a moon might not be the best for (intelligent) life due to irregular fluctuations, as systems with Jupiter`s, Saturn`s or Neptune`s close but just outside the HZ can harbour moons with life. Even acting as the opposite of an earth-moon system. (these combinations might be more common if you look at our solar system)
    If Jupiter was placed just around the `snow-line` Europa`s chances for life would be pretty good. as such the focus on earth analogues is a little unwarranted.

  • Harry R Ray October 14, 2015, 9:48

    An update on the (NOW 4 planet) Wasp 47 system. I proposed that the inner planet would turn out to be a 55 Cancri e analog. This turns out NOT to be the case. Vogt, Butler, et al have measured WASP 47′ s inner planet via radial velocity. It turns out to be MUCH MORE MASSIVE than 55 Cancri e(12.2 earths, as compared to 8.63 earths)! Assuming Spitzer’s “minima” for 55 Cancri e to be CORRECT, its density is approximately 10gcm3. Wasp 47′ INNER planet is MUCH MORE DENSE than that(approximately 20gcm3 for a radius of 1.817 earths) and could go MUCH HIGHER THAN THAT if Spitzer finds a MINIMA simalar to 55 Cancri e’s! A density SIGNIFIGANTLY ABOVE 20gcm3 would imply that is a true “chthonian” planet, the FIRST to ever be confirmed!

  • Alex Tolley October 14, 2015, 12:20

    A planet made of platinum group metals? All that wealth and so far away. :)

  • Richard October 27, 2015, 11:17

    First of all I want to thank you for this great blog. I have been a reader now for a while and always appreciate your well-written posts and insights.

    I am following the discussion about Tabby’s Star for a while now. While looking a the image above, depicting Keplers field of view, I noticed that the location of KIC 8462852 seems to be very near to the edge of the Kepler field of view. Unfortunately I couldn’t find an image acutally matching the two but I a wondering if the dimming of the star might have been caused by some bad-pixels on the edge of Kepler’s “mirror”?

    Please dont laugh at me, its just a wild guess and I am not a scientist.

    Best regards from Dresden, Germany