Spiral density waves

There was an interesting (non-climate) paper in Science, by Laura Perez and colleagues, about Spiral density waves in a young protoplanetary discs. Essentially, they used the Atacama Large Millimeter/sub-millimeter Array (ALMA) to observe a young stellar object (Elias 2-27) in the Ophiuchus star forming region and found – as shown in the image below – that this particular system had a circumstellar disc with quite prominent spiral density waves.


A great deal of my research over the last 10 years, or so, has involved studying spiral density waves in very young protostellar systems, and so I was invited to write a short Perspectives piece about this paper (I should be getting a link to a free copy shortly, so will provide it when I do). There are a number of reasons why this result is interesting. One is simply that observing structure in discs around very young stars is difficult; so it indicates that it is now becoming possible to do so. The system is about 139 parsecs (pc) away (1 pc is 206268 times the distance from the Sun to the Earth) and this image has a resolution similar to the orbit of Neptune around the Sun – this may not sound all that impressive, but the system is a long way away, and the observations are at quite long wavelengths, so it is really quite remarkable.

Another reason this is interesting, is that this disc is quite massive relative to the mass of the central protostar (the disc mass is about one-quarter that of the central protostar). This means that the spiral density waves could be due to a gravitational instability. If the disc mass is high enough, then its own gravity will try to cause the material in the disc to clump together. However, the velocity in the disc varies with distance from the central protostar, which then causes these density structures to shear out, producing spiral density waves, very like what is seen in a typical spiral galaxy (it also depends on the temperature on the disc, but this is the basic picture).

1bmqtiThe movie on the right is from one of my simulations of a massive disc around a young protostar (the properties are not the same as those of Elias 2-27, so this is just illustrative) showing the development of these spiral density waves. There are two main reasons why this is interesting. One is that these spiral density waves can act to drive mass inwards onto the central protostar, and may indeed be the main driver of mass accretion during the earliest stages of star formation. The other is that it’s fairly clear that the growth of planet building material must start when the system is very young, otherwise there wouldn’t be enough time to build the cores of the gas giant planets (Jupiter and Saturn, for example) before the disc gas dissipates. Also, there is some meteoritic evidence that differentiated objects had started forming when the Solar System was less than 1 million years old, and this means that kilometre-sized objects formed quickly (differentiation is the sinking of dense material – iron/nickel – to the centre of an object).

One issue with the growth of solids in a disc is that – if the disc is smooth and laminar – the solids should spiral inwards due to aerodynamic drag, and may fall into the central star before growing large enough to decouple from the disc gas. However, if there are any structures in the disc, then the solids can collect in these structures, rather than spiraling in to the central protostar. Therefore, in addition to driving mass accretion, these spiral density waves may also provide sites where planet building material can collect and grow. So, they may be important for both star formation and planet formation.

However, there are a number of fairly large caveats. The disc around Elias 2-27 is quite extended (i.e., it has a large outer radius) which makes a gravitational instability less likely than if it were more compact. Also, some work that myself and collaborators did about 10 years ago suggests that two-armed spirals (like that seen in the disc around Elias 2-27) are only likely in discs that have masses at least half that of the central star. The disc around Elias 2-27 is massive, but not quite that massive. So, it is possible that these spirals have a different cause. It could be embedded planets that are driving spirals out into the disc, or maybe it’s a tidal perturbation from a passing star. Even mass infall onto the disc could drive spiral density waves.

However, whatever it is, it is still an interesting result. It shows that there can be structure in discs around very young protostars and that we are now in a position to probe these structures, so as to understand how they influence both star formation and planet formation.

This entry was posted in Personal, Research, Science and tagged , , , , , , , . Bookmark the permalink.

21 Responses to Spiral density waves

  1. kevinboyce says:

    With apologies to Tom Lehrer, and hoping the formatting works…

    ALMA, tell us,
    Optical scopes are all jealous,
    Which of your resolute charms
    Got you photos of spiraling arms?

  2. Chris says:

    or in rhyming couplet form:

    beware the brickbats,
    that accompany “fairly large caveats”

  3. Chris says:


    “another stellar project,
    exploring a distant object”


  4. This comment thread is really not going as I had expected. That’s not a complaint, mind you 🙂

  5. BBD says:

    That rather wonderful simulation you have produced needs a nice EDM soundtrack.

    As for the fascinating image of Elias 2-27, I’m not hugely surprised to see spiral density waves in a protoplanetary disc as there are spiral galaxies and there seems to be a level of self-similarity in physical processes like this over a range of scales. Presumably, if the mass distribution in a galaxy is right, spiral density waves form? Or is the gravitational, rotational physics of the sorting process entirely different for spiral galaxies than for protoplanetary discs?

  6. BBD,
    For galaxies, I think there is still some disagreement about what causes spiral density waves. It could be the self-gravity of the galactic disc, it could be driven by bar structures in the centre, or it could be from interactions with other galaxies – all of these possibilities could, in fact, play some kind of role. One issue with protoplanetary discs is that their mass drops relatively quickly to the point where it becomes extremely unlikely that any structure is driven by a gravitational instability. However, the disc in Elias 2-27 is massive enough for this to be a possibility, which is one reason why this is an interesting result.

  7. Magma says:

    this disc is quite massive relative to the mass of the central protostar (the disc mass is about one-quarter that of the central protostar)

    Could a massive accretion disk be responsible for the formation of binary systems, rather than accidental capture of two independently formed stars? (Apologies for such a basic question but my knowledge of star formation physics is so out of date it would take me hours to get a useful answer for this on my own.)

  8. Magma,
    There re indications that you could objects that were massive enough to become stars in this way, so – yes – it is possible. However, it is very turbulent and dynamic environment, so it’s not likely that this is the main way in which binaries are formed. Also, if the disc were really massive, it might even not really be a disc in the sense of being a rotationally supported structure around a single central object.

  9. Dikran Marsupial says:

    I don’t really have a comment for this article, other than please can we have some more like it? ;o)

  10. Szilard says:

    DM: Ditto!

  11. izen says:

    Interesting science.

    Good to see the process of scientists looking at limited observations of a complex physical process and developing explanations of the unseen and often undetectable aspects of the events from basic theory.

    Part of the confidence in climate science is the fact that it uses the same process to reach its results in a different, but similar field. Both are attempting to understand the actions of large masses made up of individual elements within energy gradients.

    Planetary disc formation is unencumbered with the social aspects of any implications for the dominant energy use by human society however.

  12. Alex says, sorry, can you rephrase this post in the form of a quatrain please?

  13. izen says:

    There once was a planetary disc
    who’s rotation was exceedingly brisk
    With motions so gyral
    and density spiral
    It’s bulges were quite odalisque

  14. PG Antioch says:

    Forgive my ignorance of this subject, but I actually have a serious question here. AFAIK, the exact mechanism of formation of supermassive black holes at the centers of large galaxies remains unclear. Are they formed primarily by the accretion of dust & gas, by the accretion of stellar mass black hole remnants, by already-formed stars, or some combination of these? Would spiral wave theory help to distinguish between these various mechanisms? This relates to whether SMBHs form “before” their galaxies develop, “along with” them or “after” they develop.

  15. PG,
    A little beyond my direct experience. My understanding, though, is that this is still unresolved. There clearly are discs around many supermassive black holes in the centres of galaxies (active galactic nuclei) which could then drive growth of the central black hole. However, it could also be from the merger of blackholes when galaxies collide, could be the accretion of stars that form near the centres of these galaxies. I’m not sure if there is any generally accepted dominant mechanism yet.

  16. Pingback: Thinking like a planet | …and Then There's Physics

  17. John Hartz says:

    Out of curiousity, what is Richard Tols take on all of this?

    (The devil made me do it.)

  18. Pingback: Observing the earliest stages of star and planet formation | …and Then There's Physics

  19. Martin Vermeer says:

    About the similarity between these and the spiral density waves of galaxies, note that in galaxies there is star formation, with OB associations and subsequent supernova explosions that input power into the density waves, and may be driving them. Don’t think there is anything similar in circumstellar discs.

  20. Martin,
    There is some evidence that these spirals could – in some cases – be sufficiently unstable so that they break up and produce bound objects. However, these objects will probably have masses from a few times that of Jupiter, up to a fews tens of Jupiter masses (gas giant planets or brown dwarfs). Therefore, these don’t end up exploding and driving more power into the density waves. We discuss a bit of this at the end of paper (second paragraph of Section 5).

  21. Pingback: …and Then There's Physics

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.