≡ Menu

Building the Brown Dwarf Census

About a month ago we were looking at the work of Ray Jayawardhana and team on the brown dwarf 2MASS 2139, an interesting case because Jayawardhana (University of Toronto) thinks he has spotted a giant storm raging on the object, or perhaps holes in the cloud deck that allow a glimpse of deeper layers of the atmosphere within. At issue is the striking 30 percent change in brightness of the star within a mere eight hours, seeming to indicate atmospheric changes we can pick up as the brown dwarf rotates. Unlike a normal star, a brown dwarf is hot when young but gradually cools to the point where it has an atmosphere similar to that of a gas giant.

All this is part of a survey program called SONYC – Substellar Objects in Nearby Young Clusters – that uses data from the Subaru Telescope in Hawaii and the Very Large Telescope (VLT) in Chile. SONYC may change the way we look at brown dwarfs, and now it is back in the news. The latest word is that the same team has found over two dozen free-floating brown dwarfs within two young star clusters, including one object just six times more massive than Jupiter. Have a look at the image below, which depicts brown dwarfs in the cluster NGC 1333.

Image: Brown dwarfs in NGC 1333. This photograph combines optical and infrared images taken with the Subaru Telescope. Brown dwarfs newly identified by the SONYC Survey are circled in yellow, while previously known brown dwarfs are circled in white. The arrow points to the least massive brown dwarf known in NGC 1333: it is only about six times heftier than Jupiter. Credit: SONYC Team/Subaru Telescope.

What to make of the smallest of the brown dwarfs yet identified in this cluster? Aleks Scholz of the Dublin Institute of Advanced Studies comments:

“Its mass is comparable to those of giant planets, yet it doesn’t circle a star. How it formed is a mystery.”

Indeed, but the implication is that free-floating objects not a great deal larger than a Jupiter-class planet can form the same way stars form, emerging from contracting gas clouds, although there is also the possibility that some of the smaller brown dwarfs formed around a star and were later ejected from the system. SONYC is all about building a more complete census of brown dwarfs in star-forming regions to resolve questions like this, the ultimate goal being to understand how the early development of stars depends on object mass, which will in turn illuminate models of dynamical interactions and accretion as the nascent objects form. These considerations make brown dwarfs in a mass range that overlaps with massive planets a key area for research.

The SONYC census is developing our database in this area, as described in one of the two papers recently made available on this work:

We find 10 new likely brown dwarfs in this cluster, including one with a spectral type ~L3 and two more with spectral type around or later than M9. These objects have estimated masses of 0.006 to 0.02M [solar masses], the least massive objects identified thus far in this region. This demonstrates that the mass function in this cluster extends down to the Deuterium burning limit and beyond. By combining the findings from our SONYC survey with results published by other groups, we compile a sample of 51 objects with spectral types of M5 or later in this cluster, more than half of them found by SONYC. About 30-40 of them are likely to be substellar.

The astronomers studied both NGC 1333 and the Rho Ophiuchi star cluster with Subaru at both optical and infrared wavelengths. NGC 1333, a young cluster thought to be no more than a million years old, turns out to have a higher population of brown dwarfs than the average young cluster, a fact that may offer clues to different conditions within the cluster that affect brown dwarf formation. The cluster houses half as many brown dwarfs as normal stars.

The paper continues:

The star vs. brown dwarf ratio in NGC1333 is significantly lower than in other nearby star forming regions, possibly indicating environmental differences in the formation of brown dwarfs. We show that the spatial distribution of brown dwarfs in NGC1333 closely follows the distribution of the stars in the cluster. The disk fraction in the brown dwarf sample is < 66%, lower than for the stellar members, but comparable to the brown dwarf disk fraction in 2-3 Myr old regions. The substellar members in NGC1333 show a large fraction of highly flared disks, evidence for the early evolutionary state of the cluster.

The paper cited above is Scholz et al., “Substellar Objects in Nearby Young Clusters (SONYC) IV: A census of very low mass objects in NGC1333,” accepted for publication in the Astrophysical Journal (preprint). The second paper on this work is Muzic et al., “Substellar Objects in Nearby Young Clusters (SONYC) V: New brown dwarfs in rho Ophiuchi,” also accepted by the Astrophysical Journal (preprint).


Comments on this entry are closed.

  • jkittle October 17, 2011, 13:44

    this is fantastic work and will help decipher not only the Brown dwarf populations but also open a window on stellar formation in general. The authors are working with within the constraints of having a thick, warm and wet atmosphere above them. Space based telescopes like the proposed WFIRST will have a better view, particularly if they have large reflectors ( greater than 1 meter) look deeper into the infrared than ground based scopes can manage ( For example from 2 to 8 microns) and have decent field of view ( about a 4 degrees squared or so) The JWST is a great scope as far as resolution and sensitivity, but has limited field of view and cannot be used to survey much of the sky. It will be a great instrument to zoom in on the brown dwarfs other scopes discover. Even the Subaru ‘scope is limited in looking at cool and objects because of the limits of wavelength.

  • andy October 17, 2011, 14:00

    6 Jupiter masses? Guess this is yet more evidence that deuterium fusion should not be used to distinguish brown dwarfs from planets.

  • Martin J Sallberg October 17, 2011, 14:49

    Is it possible that a not-quite-sharp limit of Deuterium fusion means that factors other than mass can determine whether or not it becomes a brown dwarf in a small mass range just at the limit? That could mean that more precise observation than mere mass measurement may be necessary to determine whether or not it really is a brown dwarf.

  • Holger October 17, 2011, 15:24

    could you clarify on the definition of “brown dwarf” you (and the article you cite) use? A 6 Jupiter-mass object is not expected to be able to fuse deuterium, so I’d call it a sub-brown dwarf (see e.g. Wikipedia definition) rather than a “proper” brown dwarf.

  • Paul Gilster October 17, 2011, 15:40

    Holger, here’s the brown dwarf definition cited in the paper:

    Brown dwarfs are objects with masses too low to sustain stable hydrogen burning… and as such intermediate in mass between low-mass stars and giant planets…

    And then note this:

    Surveys in star forming regions indicate that the mass regime of free-floating brown dwarfs extends down to masses below the Deuterium burning limit at 0.015 [solar masses].”

    The findings do bring the deuterium-burning issue to the fore, as andy mentions above.

  • jkittle October 17, 2011, 15:57

    Holger. I hold with the deuterium fusion limit as the definition of the low mass limit for brown dwarfs. Neptune and even earth has gasses and is “sub stellar”
    It turns out that the density function is driven a lot by the composition of the Dwarf, with much higher densities possible if heavier atoms are present in the protoplanet cloud. The more compact objects may support fusion at somewhat lower masses that a BD made from more primordial stuff. .. so the fusion limit is not just a function of mass.

  • Gregory Benford October 17, 2011, 19:03

    Is there a probability estimate of finding a brown dwarf closer than Alpha Centauri?

  • Paul Gilster October 17, 2011, 21:35

    Good question. I know of no such estimate but I suspect Adam Crowl may know of anything in the literature. Adam?

  • Joy October 18, 2011, 3:14

    Any BDs near Earth would no longer have heat of recent formation, being old and cold, they would have very low luminosity and could be very hard to detect. Still, detection becomes less likely with every negative finding from IR surveys by space observatories. A decade ago, I would have guessed odds up to 50/50, now I would guess odds less than 10%

  • Holger October 18, 2011, 6:13

    thanks for your answer. But this “definition” is obviously incomplete (maybe meant to be informal), since giant planets (and even Earth) are also “objects with masses too low to sustain stable hydrogen burning”. Some criterion for distinguishing brown dwarfs from planets is needed – the two I have seen used are “deuterium-burning” or “formed from a gas cloud like a star”. From your second quote, it seems the latter one is used. I.e. some of these objects are actually “sub-brown dwarfs”, intrinsically undistinguishable from same-sized giant planets.

    I agree, I’d also prefer the “deuterium fusion” lower limit being used for brown dwarfs. But is there any combination of materials resulting in a 6-Jupiter-mass sized object burning deuterium?

  • Ronald October 18, 2011, 6:14

    “brown dwarfs in a mass range that overlaps with massive planets”

    This makes me wonder, like others here, what distinction remains between giant gas planets and brown dwarf stars. According to the Extrasolar Planets Encyclopedia (exoplanet.eu) many found planets (i.e. orbiting stars) have masses considerably greater than 6 Mj, but then some of these could be brown dwarfs in a binary stellar system.

    So, is there simply a (mass) continuum from planets via brown dwarfs to ‘real’ (hydrogen fusing) stars? Or is there a fundamental distinction between two different categories, based on their formation and reflected in their composition?
    In the latter case, I would think that the distinction is that stars form from condensing hydrogen gas clouds and (giant) planets form by secondary accretion around a rocky core.

  • ljk October 18, 2011, 8:14

    Man, are there a lot of Worldships and Jupiter Brains out there!

  • Paul Gilster October 19, 2011, 8:22

    The odds are pretty high IMO. Lorenzo Iorio’s work on perturbations of Saturn seem to indicate something is out there, roughly where Matese & Whitmire have posited their “Tyche”, so I suspect the Sun has a brown dwarf companion. Wide separation binary surveys are picking up companions out ~2,500 AU or so. Tyche would fit that picture.

    OTOH field brown dwarf estimates are a lot lower than the old single exponent Initial Mass Function led some to suppose. The low-mass “knee” surprised people when it showed up in globular clusters and it does seem to hold amongst disk stars too. The old estimate suggested ~x10-x100 brown dwarfs to regular stars, but current surveys suggest about 1-to-1 numbers.

    But we have a mass of WISE data being combed through, so maybe we’ll know before long.


  • kzb October 19, 2011, 13:36

    Came across this little presentation by accident yesterday. It is about a search for “Tyche” using the recent WISE data release. Any comments?

    Searching the WISE preliminary catalog for massive planets in the Oort Cloud
    Matese & Whitmore