Planet Formation around TRAPPIST-1

by Paul Gilster on June 9, 2017

Just how did the seven planets around TRAPPIST-1 form? This is a system with seven worlds each more or less the size of the Earth orbiting a small red dwarf. If these planets formed in situ, an unusually dense disk would have been required, making planet migration the more likely model. But if the planets migrated from beyond the snowline, how do we explain their predominantly rocky composition? And what mechanisms are at work in this system to produce seven planets all of approximately the same size?

New work out of the University of Amsterdam attempts to resolve the question through a different take on planet formation, one that involves the migration not of planets but planetary building blocks in the form of millimeter to centimeter-sized particles. Chris Ormel (University of Amsterdam) and team note that thermal emission from pebbles like these has been observed around other low-mass stars and even brown dwarfs. The researchers believe these migrating particles become planetary embryos as they reach the snowline, which at TRAPPIST-1 occurs at about 0.1 AU.

Once within the snowline, the embryos would grow by the accretion of rocky pebbles from the inner circumstellar disk, with inward migration eventually stopping at the inner edge of the disk. A key assumption here is that the planets of TRAPPIST-1 formed sequentially rather than simultaneously, a novel concept indeed. So let me go to the paper at this point:

In our model we assume that the H2O iceline is the location where the midplane solids-to-gas ratio exceeds unity, triggering streaming instabilities and spawning the formation of planetesimals. These planetesimals merge into a planetary embryo, whose growth is aided by icy pebble accretion. Once its mass becomes sufficiently large, it migrates interior to the H2O iceline by type I migration, where it continues to accrete (now dry) pebbles until it reaches the pebble isolation mass.

The process then begins again for a second planet:

After some time, a second embryo forms at the snowline, which follows a similar evolutionary path as its predecessor. Even though the inner planet’s growth could be reduced by its younger siblings’ appetite for pebbles, it always remains ahead in terms of mass. Planet migration stalls at the inner disk edge, where the planets are trapped in resonance.


Image: Astronomers from the University of Amsterdam (the Netherlands) present a new model for how seven earth-sized planets could have been formed in the planetary system Trappist-1. The crux is at the line where ice changes to water. Credit and copyright: NASA/R. Hurt/T. Pyle. And please note this JPL news release on the artists who produced this image. All too often, artists like Tim Pyle and Robert Hurt receive scant attention in the stories that run their work. It’s excellent to see their background and methods explained.

The TRAPPIST-1 planets, indeed, form what the authors call ‘a resonant convoy,’ with the outer planets ‘pushing’ on the inner ones. The paper’s numerical simulations produce the observed planetary system with the exception that a 3:2 mean motion resonance emerges among planets b and c, as well as among c and d. Although neither pair is presently at the 3:2 MMR, the authors argue that during the disk dispersion phase of the system’s formation, the 3:2 MMRs of these pairs were broken, leaving us with the overall architecture we see today.

The paper’s most radical contention is that planets have assembled at a specific location, the snowline, as opposed to forming in situ or migrating from their formation regions beyond the snowline. Clearly, many questions remain, including how the streaming instabilities induced at the snowline operate in the presence of planetary embryos. The paper does, however, make a prediction: If a giant planet forms rapidly at the snowline, it should end the flux of pebbles to the inner disk, depriving it of planet-building material. From the paper:

Hence, we expect a dichotomy: when giant planet formation fails, pebbles can drift across the iceline to aid the growth of super-Earths and mini-Neptunes. Conversely, when a giant planet forms at the iceline we expect a dearth of planetary building blocks in the inner disk. Therefore, the close-in super-Earth population found by Kepler and the cold Jupiter populations found chiefly by radial velocity surveys should be anti-correlated – a prediction that could be tested with future exoplanet surveys.

The paper is Ormel et al., “Formation of Trappist-1 and other compact systems,” accepted at Astronomy & Astrophysics (abstract). A preprint is available, but be aware that a number of internal references are not yet filled in, another reason not to assume that preprints necessarily mirror the final paper.


{ 13 comments… read them below or add one }

Alexander Tolley June 9, 2017 at 11:40

I like that this hypothesis is testable. This could be determined quite quickly.

There seems to be quite a number of different planetary formation theories. How many are mutually exclusive and how many can be complementary models depending on context?


Robert Feyerharm June 9, 2017 at 14:37

The New York Times published graphics of the extrasolar systems discovered by the Kepler Telescope in 2015 (see, and it’s immediately evident there is a dichotomy between systems with hot Jupiters in the top rows vs. multiple super-Earths/Neptunes, with few planetary systems resembling our own.

BTW Tau Ceti is also suspected to be a Trappist 1-like system, with 5 super Earths orbiting within 1.5 AU.


Ronald June 9, 2017 at 17:54

So, a trichotomy. The hot Jupiter systems are relatively rare. The compact systems of medium-sized planets (gas-dwarfs, ice-giants) are very common. Our kind of open system seems rather rare, but part of that may be observational bias.


Sanford Zane Meschkow June 9, 2017 at 17:27

Robert, where can one find information on the probable size of the possible Tau Ceti super Earths? Anything that might be habitable?


Robert Feyerharm June 12, 2017 at 0:29

Here’s a link to the paper describing the (possible) discovery of five planets orbiting Tau Ceti:

The researchers estimated minimum masses for the planets in the 2.0 – 6.6 Earths range, we can only speculate on the diameters of the planets. The Doppler signals for the planets are very weak, perhaps due to the fact that the star Tau Ceti is oriented with it’s rotational axis pointing towards Earth and therefore we’re seeing its planetary system nearly face on.


Ronald June 12, 2017 at 8:26

Remarkably, does not include Tau Ceti (?).
You can find some interesting information in this publication by Tuomi et al.:
And this one:

And about possible habitable planets on this site:


Geoffrey Hillend June 9, 2017 at 18:07

I am skeptical about the idea that the planets had to form one at a time from the snow line. It certainly is possible but it might be unlikely. In situ seems more logical.

Principles that apply to a large solar system might not apply to a small system since a large system is much more spread out than a small one due to the lesser gravity of the small one like a red dwarf. The lower gravity keeps the system size much smaller. Consequently, the lack of planets of high densities beyond a snow line position and inside it is normal due to everything coming from the same gaseous nebula in a much smaller and closer distance from the protostar. In otherwords, the planets might not have to had migrated, but did come from a small, dense cloud.

The anti-correlation between Jupiter’s and super Earths is due to the small size of the red dwarf star solar system. The gravity of the Jupiter would use up the gaseous nebula so that you would not get many or any super Earths in that system which explains the anti-correlation.


Geoffrey Hillend June 9, 2017 at 18:25

Excuse me. I should have wrote: Consequently, the lack of planets of high densities beyond the snow line and majority of them inside it is due to the small size of the gaseous nebula.


ljk June 10, 2017 at 13:23

The art of exoplanets:

To quote:

“For the public, the value of this is not just giving them a picture of something somebody made up,” said Douglas Hudgins, a program scientist for the Exoplanet Exploration Program at NASA Headquarters in Washington. “These are real, educated guesses of how something might look to human beings. An image is worth a thousand words.”


Geoffrey Hillend June 10, 2017 at 17:05

I am assuming that they mean the snowline in this theory is the distance where water must be frozen which is just outside the life belt where liquid water can be stable and not frozen. I don’t see why rocky planets would be limited to forming there and have to migrate.

The accepted mainstream astrophysical theory of gaseous or diffuse nebula or stellar nurseries is that planetesimals and the protostar form at the same time so the thermal emission could not be a factor until a star is born. Also, gravity is stronger than thermal emission in the attraction and accumulation of dust and gas in the formation of a planetesimal.


Michael June 11, 2017 at 3:05

The snowline for Red dwarfs is a little bit more complicated by the fact they have long contraction phases which pushes the snowline much further out. So we could have loads of waterless worlds of our gravity and water compliment in the habitable zones when it is on the main sequence.


Ronald June 12, 2017 at 3:10

And a new study on the composition of the Trappist planets;
“B. Quarles, E.V. Quintana, E.D. Lopez, J.E. Schlieder, T.Barclay. Plausible Compositions of the Seven TRAPPIST-1 Planets Using Long-term Dynamical Simulations. Astrophysical Journal Letters, June 2017”.


Geoffrey Hillend June 12, 2017 at 16:12

We will have to wait a very long time before that snowline changes and for TRAPPIST-1 to contract since a red dwarf star with only 8 percent the mass of our Sun does not move off the main sequence for trillions of years.

TRAPPIST-1 at only 8 percent the mass of our Sun is not massive enough to cause a gravitational compression high enough to for the fusion of Helium so it does not have a red giant phase. It goes straight to the electron degeneracy of a white dwarf, a process that takes trillions of years since less massive stars burn fuel at a slower rate due to less gravitational contraction.


Leave a Comment

{ 2 trackbacks }