Title: Jupiter’s Decisive Role t’sn the Inner Solar System’s Early Evolution
Authors: Konstantin Batygin and Gregory Laughlin
First Author Institution: Division of Geological and Planetary Sciences, California Institute of Technology, 1200 E. California Blvd., Pasadena, CA 91125, USA
Status: Submitted to Proceedings of the National Academy of Sciences of the United States of America
Since the discovery of the first extra-solar planet around another star in 1995 we now know of ~ 4000 candidate planets in our galaxy. All of these discoveries have been key in improving our understanding of the formation of both planets and star systems as a whole. However a full explanation of the processes which form star systems is still elusive, including a description of the formation of our very own Solar System.
The problem with all these exoplanets and star systems we’ve discovered so far is that they suggest that the Solar System is just weird. Most other systems seem to have massive planets similar to the size and mass of Neptune but which are the distance of Mercury to the Sun from their own star (often these planets can be as big as Jupiter – since these are the easiest for us to detect). For example the famous Kepler-11 system is extremely compact, with 6 planets with a total mass of about 40 Earth masses all within 0.5 AU (astronomical unit – the distance of the Earth from the Sun) orbiting around a G-type star not at all dissimilar from the Sun.
Figure 1 shows all the Kepler detected planets with masses less than Jupiter within the orbit of Mars from their own star. So if most other star systems seem to be planet mass heavy close into their star – why is the Solar System so mass poor and the Sun so alone in the centre?
The authors of this paper use simulations of how the orbital parameters of different objects in systems change due to the influence of other objects, to test the idea that Jupiter could have migrated inwards from the initial place it formed to somewhere between the orbits of Mars and Earth (~ 1.5 AU). The formation of Saturn, during Jupiter’s migration, is thought to have had a massive gravitational influence on Jupiter and consequently pulled it back out to its present day position.
If we think first about how star systems form, the most popular theory is the core-accretion theory, where material around a star condenses into a protoplanetary disc from which planets form from the bottom up. Small grains of dust collide and stick together forming small rocks, then in turn planetesimals and so on until a planet sized mass is formed. So we can imagine Jupiter encountering an army of planetesimals as it migrated inwards. The gravitational effects, perturbations and resonances between the orbits of the planetesimals and Jupiter ultimately work to cause the planetesimals to migrate inwards towards the Sun. The simulations in this paper show that with some simplifying assumptions the total amount of mass that could be swept up and inwards towards the Sun by Jupiter is ~10-20 Earth masses.
Not only are the orbital periods of these planetesimals affected but their orbital eccentricities (how far from circular the orbit is) are also increased. This means that within that army of planetesimals there’s now alot more occasions where two orbits might cross initiating the inevitable cascade of collisions which grind down each planetesimal into smaller and smaller chunks over time. Figure 2 shows how the simulations predict this for planetesimals as Jupiter migrates inwards.
Given the large impact frequency expected in a rather old protoplanetary disc where Jupiter and Saturn have already formed, the simulations suggest that a large fraction, if not all, of the planetesimals affected by Jupiter will quickly fall inwards to the Sun, especially after Jupiter reverses its migration direction. This decay in the orbits is shown in Figure 3 with each planetesimal getting steadily closer to the Sun until they are consumed by it.
The orbital wanderings of Jupiter inferred from these simulations might explain the lack of present-day high mass planets close to the Sun. The planetesimals that survived the collisions and inwards migration may have been few and far between, only being able to coalesce to form smaller rocky planets like Earth.
The next step for this theory is to test it, on another star system similar to our own with giant planets with orbital periods exceeding 100 days. However our catalogue of exoplanets is not complete enough to provide such a test yet. Finding these large planets at such large radii from their star is difficult because their long orbital periods coincide with how often we have the chance of observing a transit. For example if we wanted to detect a Neptune-like planet in a Neptune-like orbit, a transit would only occur every 165 years. Also, detecting small planets close to a star is also difficult as the current telescope sensitivities don’t allow us to detect the change in the light of a star for planets so small.
So perhaps we just haven’t been looking long enough or with good enough equipment to find star systems like ours. However with missions like GAIA, TESS and K2 in the near future perhaps we’ll find that the Solar System is maybe not as unique as we think.
Great post! I had never considered the idea that the formation of other planets could affect the position of others to such a wide extent. It will be interesting to see how our theories develop with more data from the new missions.
Thanks for this post! You note that we’ve discovered a lot of planets the size of Jupiter because those are easier to detect — is it possible that we have trouble finding planets that are less massive than Jupiter rather than that so few exist?
Hi Anne! Yes I discuss that at the end of the post that it could just be detection bias. The hope is that with future missions we should be able to figure this out.
Love learning about the theories behind the formation of our solar system! I don’t think I’ve heard of Jupiter reversing it’s migration direction- is this a normal occurrence during the formation of larger planets like Jupiter?
Great post Becky… Interesting read. I would like to know if you are working from the Nice model first proposed in 2005.
Awesome. I’ve always wondered why we Jupiter isn’t a hot jupiter in our solar system.
“The problem with all these exoplanets and star systems we’ve discovered so far is that they suggest that the Solar System is just weird.”
I don’t see how they suggest that our system is weird. The data doesn’t yet cover systems of our size and small planets as some of ours to any degree of statistical significance, I think.
The data so far seem to say that:
a) systems are very individual
b) systems have typically 0-1 planets or 4-8 planets.
Our system checks out as normal on both accounts.
This doesn’t say that this work is wrong, only that it isn’t as motivated by observed constraints as the Nice models are.
@Jennifer Shi: In my layman naivety I thought the old Nice models had looked at this, and a very quick check seems to suggest they have:
“In the case of the original Nice model, the slow approach of Jupiter and Saturn to their mutual 2:1 resonance, necessary to match the timing of the Late Heavy Bombardment, can result in the ejection of Mars and the destabilization of the inner Solar System. A step-wise separation of Jupiter’s and Saturn’s orbits due to gravitational encounters with one of the ice giants, called the jumping-Jupiter scenario, has been shown to be necessary to avoid these issues. The frequent ejection of the ice giant encountering Jupiter has led some to propose an early Solar System with five giant planets, one of which was ejected during the instability.”
[ http://en.wikipedia.org/wiki/Nice_model ]
I assume the operative word is “can”, and that the modification covers the remaining cases (and a 5th later ejected giant would help make Uranus the correct size as I remember it), but I haven’t yet looked into those references.
It’s amazing to think that we’ve accomplished so much in the search for exoplanets in only a few years relative to all of astronomy and science. I am sure that with more data on other exoplanet systems, we’ll soon know with relative surety how our solar system formed.
Is there any room in this theory for Saturn to have been formed first? Or has that been excluded by the simulations?
You mentioned that we may have yet to discover solar systems like ours. How biased are the results we see now given that large, close orbiting planets are easy to detect? Is it possible that solar systems like ours are quite common and we just haven’t detected them, or has that possibility been somewhat eliminated by what we see?
How novel an idea is it that Jupiter may have once migrated in, only to be pulled back out by Saturn? It isn’t something I’ve heard of before, but here it seems like something people at least consider plausible. Is there any observational evidence for something like this? Do we tend to see fewer hot Jupiters in systems with more than one large planet, for instance?
I am with Sam and was greatly surprised by this theory involving Jupiter migrating first closer and then further from the sun. How accurate is this theory? Great post by the way!
Do these findings leave room for other possibilities besides the grand tack model of Saturn pulling Jupiter outwards? If not, is this model now generally accepted or are there still critics in the scientific community?
I never knew Jupiter would affect Saturn’s position in such a way. I wonder if tidal interactions between Jupiter and Uranus caused Uranus to have it heavily tilted axis
What are other particularities of our solar system that we do not observe in the majority of others whose existence might explain this problem as well? As in: the cause of their existence, which might be more easy to determine, might also be the cause of this problem’s existence.