Stellar activity is typically attributed to magnetic field interactions, but in brown dwarfs this emission is more of a mystery.

For the most part, brown dwarfs don’t do very much; they just sit there and slowly cool off.  But a rare few are “hyperactive”, exhibiting emission lines that both vary and persist over long periods of time. The source of this activity has been a mystery for nearly a decade, but new observations we’ve conducted with the IRTF, Magellan and Keck Telescopes show that one of these hyperactive dwarfs has a very faint companion, and that it is hyperactive because it is very old and relatively massive.

Strong hydrogen alpha emission is seen at the left in this optical spectrum of 2MASS 1315-2649 obtained at the Magellan Telescopes. The nature of this persistent emission has been a mystery for nearly a decade.

The hyperactive dwarf in question is 2MASS J13153094-2649513 (2MASS 1315-2649 for short), an L-type dwarf independently discovered in 2002 by John Gizis and Pat Hall in the Two Micron All Sky Survey.  Both researchers were surprised to find very strong hydrogen alpha emission coming from this source, particularly since very few L dwarfs show any such emission.   As the emission was seen to persist, various ideas were proposed to explain it, such as accretion from a disk or a close companion, an unusually strong magnetic field, or other heating mechanisms. However, subsequent observations were unable to firmly confirm or refute these hypotheses.

Laser Guide Star Adaptive Optics images of 2MASS 1315-2649, revealing its faint companion to the southeast. The separation of the two sources is about 1/3 of an arcsecond, or 0.0001 degrees.

Is this companion responsible for 2MASS 1315-2649′s hyperactivity?  Probably not, as the two sources are about 6.6 Astronomical Units (AU) away from each other, too far apart to really interact.  Yet the fact that the companion is so faint gives us another clue.  It turns out that most brown dwarf binaries are nearly equal in mass, a characteristic that might be related to how they form.  If that trend applies here, then the 2MASS 1315-2649 system would have to be very old – of order 5-10 billion years old – to give the two sources time to diverge in brightness.  That would make the brighter source in 2MASS 1315-2649 more massive, perhaps massive enough to be a hydrogen-burning star rather than a brown dwarf.  This turns out to be an important point, as a recent study by Ulrich Christensen finds that more massive brown dwarfs have stronger magnetic fields.  From this line of reasoning, it would appear that strong magnetic fields might the cause of 2MASS 1315-2649′s hyperactivity.

An image of WISE 1906+4011 from the Kepler Observatory; the green mask marks the location of this faint source.

The Kepler Observatory is uncovering numerous planets in close orbits around distant stars.  Now it has a brown dwarf under its belt.  A search for nearby cool dwarfs by University of Delaware collaborator John Gizis uncovered the first L dwarf in Kepler, a source that may teach us much about clouds in cool brown dwarfs.

The new source, WISEP J190648.47+401106.8 (or WISE 1906+4011 for short) was identified by Dr. Gizis in a search for nearby late-type M and L dwarfs with high proper motions, that were detected by the Wide-field Infrared Survey Explorer (WISE) and the Two Micron All Sky Survey (2MASS).  Of the hundreds of fast-moving stars, WISE 1906+4011 stood out both for its fast motion (0.48 arcsecond/year, the equivalent of 0.13 degrees/millenium) and the fact that it lies in the footprint of the Kepler Observatory.

Near-infrared spectrum of WISE 1906+4011 (black), compared to a template dwarf (red), confirming this source to be an early-type L dwarf.

The colors of WISE 1906+4011 indicated a very red, low-temperature object, and a near-infrared spectrum obtained with the SpeX spectrograph on the NASA Infrared Telescope Facility confirmed it as an L-type dwarf at a distance of about 54 light-years.  It is the first L dwarf to be identified in the Kepler field, and is now the target of a year-long monitoring campaign. Kepler’s sensitivity and stability has made it a planet-finding machine, with over 1000 candidates and 20 confirmed planets identified so far.  We hope to exploit that sensitivity to explore brightness variations induced by revolving mineral clouds in the atmosphere WISE 1906+4011, and thereby learn about weather patterns on brown dwarfs.

This result will be published in in the Astrophysical Journal Letters; a preprint is available here.  The authors are John Gizis (U. Delware), Nicholas Troup (U. Delaware) and Adam J. Burgasser (UCSD).

In this WISE false-color image, a cold brown dwarf appears green

The Folded Port Infrared Echellette (FIRE) spectrograph, commissioned in March 2010, is contributing to the discovery of some of the coldest and least luminous brown dwarfs found to date.  In a paper accepted for publication to the Astrophysical Journal, we report the discovery of five late-type T dwarfs from the WISE survey identified by FIRE.

The Wide-field Infrared Survey Explorer, or WISE, has recently imaged the entire sky at mid-infrared wavelengths, between 3 and 22 µm, beyond what our eyes can discern.  Yet it is at these wavelengths that cold brown dwarfs emit the majority of their light. With specially-selected filters, WISE can both detect and discriminate the coldest brown dwarfs, by measuring the absorption caused by methane present in their atmospheres.  When an object of the right color shows up, it is a good candidate to be a very cold brown dwarf.

Near-infrared FIRE spectra of five cold brown dwarfs discovered by WISE. The spectral types are listed on the right, while the labels indicate what gases are creating the absorption features seen in these spectra.

Confirming these candidates requires spectroscopy, however, and this is a challenge when one is trying to detect very cold, very faint brown dwarfs.  Fortunately, FIRE was designed to be highly sensitive in its low-resolution prism mode, and we have used this mode to identify and characterize five new, very late-type T dwarfs (these are the class of brown dwarfs that harbor methane in their atmospheres).  The discoveries have temperatures between 600 K and 900 K, or 620 F to 1160 F.  While this seems hot from a human perspective, these brown dwarfs are up to 10 times cooler than the Sun, and only twice as warm as the Earth’s atmosphere (on a warm summer day).

This result will be published in in the Astrophysical Journal.  The lead authors is Adam J. Burgasser (UCSD).

The newly discovered 300 K brown dwarf GJ 3483B, seen in the mid-infrared by Spitzer (left) but invisible in the near-infrared (right).

In collaboration with Penn State colleagues Kevin Luhman and John Bochanski, we have uncovered a remarkably cold brown dwarf whose atmosphere appears to be a mere 300 K.  This is the equivalent of a warm day at the beach, not what one expects from the surface of a star.  In fact, the properties of this brown dwarf are so unusual that it may actually be a room temperature superplanet on an incredibly long orbit around a dead star.

The source, GJ 3483B, is a co-moving companion to the nearby white dwarf GJ 3483 (aka WD 0806-661), a DQ4 at a distance of 19.2 pc (62.6 lightyears).  It was detected by the Spitzer Space Telescope in the mid-infrared 4.5 µm band in 2004 and again in 2009, and the motion of the brown dwarf agrees exactly with that of the white dwarf, validating its physical association.  Because we know the distance to the white dwarf, we are able to measure the absolute brightness of  GJ 3483B in the mid-infrared. It turns out to be about 2.5 magnitudes, or a factor of 10 times, fainter than the coldest brown dwarf known at the time, UGPS 0722-05.  Moreover, it has been undetectable in several other photometric bands to strict limits, reaffirming its very dim nature.  Using the mid-infrared brightness, the age of the white dwarf (about 1.5 billion years – a third of the age of our Sun) and brown dwarf evolutionary models, we find an estimated atmospheric temperature of 300 K – the same as the Earth – and a mass about 7 times that of Jupiter.

Absolute near-infrared brightnesses of known brown dwarfs versus color. The previous record holder, UGPS 0722-05, is shown in green; the recently discovered cool binary, CFBDSIR 1458+10AB, is indicated in red. The best limit for GJ 3483B (downward arrows) is roughly 10 times fainter.

The temperature, mass and association of GJ 3483B with another star makes it seem very planet-like in appearance.  However, it has an incredibly wide orbit – about 2500 Astronomical Units (1 AU is equal to the Earth-Sun distance) – implying an orbit period of at least 150,000 years.  Is this a planet that has been pushed into a wide orbit after its host star, originally twice the mass of the Sun, shed 70% of its mass at the end of its stellar lifetime? Or is this simply a double star system, with one star happening to have a very low, planetary-like mass?

The low temperature inferred for GJ 3483B is itself remarkable, as the only equivalent-temperature atmosphere we know of is that of Earth.  It is so cold that ice clouds are expected to be present in its upper atmosphere – an ice star!  However, GJ 3483B would not itself be “habitable” – its hydrogen- and helium-rich atmosphere would be poisonous to life as we know it.

GJ 3483B is probably not the only one of its kind.  Similarly cool brown dwarfs – referred to as “Y dwarfs” – are being found as companions to other brown dwarfs, and as isolated objects by NASA’s WISE mission.  Yet GJ 3483B remains the coldest brown dwarf found so far, and studying its dim atmosphere will require the next generation of large ground-based and space-based telescopes.

This result was published in in the Astrophysical Journal Letters.  The authors are Kevin Luhman (Penn State), Adam J. Burgasser (UCSD), and John Bochanski (Penn State).  Also see the Science News & Analysis article describing the discovery and this article from Sol Station.

Thanks to Genevive Bjorn for help setting this up.  Want to learn how to make your own group page on Facebook and linking up your research blog posts to your twitterfeed? Check out http://www.facebook.com/pages/create.php and http://twitterfeed.com/.

The system, named 2MASS J01303563-4445411, was discovered serendipitously with the SpeX imager/spectrometer, mounted on the 3m NASA Infrared Telescope Facility.  The primary of the system, an M9 dwarf, was originally targeted in early December 2009 as part of an ongoing near-infrared spectroscopic survey to populate the SpeX Prism Spectral Libraries survey.  However, an image of the field revealed a faint companion  about 3.3″ (0.0009 degrees) to the east.  A few weeks later, U. Hawaii graduate student Dagny Looper obtained a spectrum of the companion and found it to be consistent with that of an L6 dwarf.  The estimated distances of the two stars were in agreement with each other, and a statistical technique developed by Mr. Dhital as part of his PhD thesis research (Sloan Low-mass Wide Pairs of Kinematically Equivalent Systems, or SLoWPoKES) indicate that they are likely to be a physically bound pair.

Near-infrared spectra of the 2MASS J01303563-4445411 primary (top) and secondary (bottom), taken with SpeX

However, the system, which probably consists of a very low mass, hydrogen-fusing star and a non-fusing brown dwarf, are not bound particularly strongly.  With a projected separation of 130 AU, the gravitational force between them is about the same as that between the Sun and Uranus, and one orbit takes roughly 4000 years to complete (monitoring this orbit is not an ideal project for a PhD thesis).  Such weakly-bound systems were believed to be ripped apart by encounters with other stars, as most low-mass binary orbits are found to be 10 times smaller, with the exception of a handful of very young binaries which may not have had time to be disrupted.  2MASS J01303563-4445411, on the other hand, appears to be a relatively old system. Developing theories describing the formation of low-mass stars and brown dwarfs must therefore account for the creation of such wide, fragile pairs.

This result was published in in the Astronomical Journal.  The authors are Saurav Dhital (Vanderbilt University), Adam J. Burgasser (UCSD), Dagny L. Looper (U. Hawaii) and Keivan G. Stassun (Vanderbilt University).

A UKIDSS false color image of the Ross 458 system, composed of a pair of M dwarfs (bright star in upper left corner) and the planetary-mass brown dwarf Ross 458C (circled in lower right corner).

To study the atmospheres of young planets outside our Solar System, we need not look  far.  The first brown dwarf science with the newly-commissioned FIRE spectrograph has revealed the presence of rock clouds in the atmosphere of a planetary-mass companion to the nearby Ross 458 system.  The presence of these clouds, and the planetary nature of the source, defy prior expectations.

The source in question is Ross 458C, a brown dwarf candidate identified in the UKIDSS survey in early 2010 by two independent studies led by Drs. Rolf-Dieter Scholz and Betrand Goldman.  This source, also known as ULAS J130041.72+122114.7, is located 1.7′ (0.028 degrees) southeast of the Ross 458 system, a pair of magnetically active M dwarfs only 11.2 pc (36.5 light-years) from the Sun. The colors and faintness of Ross 458C, and that fact that it co-moves with the Ross 458 system, led both studies to conclude that it was potentially a very cool and very low-mass brown dwarf companion.  However, neither study had the necessary spectral data to probe its atmosphere.

Enter the Folded-port Infrared Echellette, or FIRE, spectrograph, which was commissioned on the Magellan 6.5m Baade telescope in late February 2010 (see this prior blog post ).  One of the first targets we aimed FIRE at was Ross 458C, in order to determine the nature of this interesting faint object.  The resulting spectrum revealed strong bands of water and methane gas, typical of low-temperature brown dwarfs of the T spectral class.  We classified Ross 458C as a T8 dwarf, found its luminosity to be 400,000 times fainter than the Sun, and estimated its surface temperature to be a mere 650 °K (710 °F) – about the temperature of a pizza oven.  The rapid rotation and strong magnetic activity of Ross 458A indicates that the system is relatively young – a mere 150-800 million years old – which would give Ross 458C a mass of only 6-11 times that of Jupiter based on evolutionary models.  This is similar to the masses of the planets found orbiting the HR 8799 system.

FIRE spectrum of Ross 458C (black line) compared to best-fit atmosphere models with (blue line) and without (red line) clouds. The cloudy model provides a better fit, evidence that clouds are present in the atmosphere of this source.

Analysis of the FIRE spectrum of Ross 458C revealed another surprise – its atmosphere appears to be covered in clouds.  Comparing the spectrum to models created by Drs. Mark Marley and Didier Saumon, we found that the best fits could be made only when the absorption effects of clouds were included in the models.  The presence of these clouds – which are made of rocks and metal rather than water – was unexpected, as it is has long been assumed that such clouds are sunk deep below the atmospheres of T-type brown dwarfs. However, the young age of Ross 458C may allow these clouds to reside at higher layers in the atmosphere and form larger particles, in both cases making them easier to detect.

Ross 458C not only challenges our understanding of clouds in brown dwarf atmospheres, it challenges our very definition of what a brown dwarf is.  As a non-fusing object in orbit around a pair of stars, and with a mass below the limit of even deuterium fusion (about 13 Jupiter masses), it satisfies all of the requirements for a “planet” as laid out by the International Astronomical Union in 2006 and in an article published by Drs. Gibor Basri and Mike Brown, also in 2006.  Yet this is certainly a very unusual planet, since it has an orbit that is roughly 1000 AU in radius and is at least 6 times more massive and hotter than Jupiter.  As in many scientific endeavors, the more we learn, the more we find our assumptions and expectations might be wrong!

This result was published in in the Astrophysical Journal, and presented at the 16th Cambridge Workshop on Cool Stars, Stellar Systems and the Sun and the 217th American Astronomical Society Meeting. Authors include Adam J. Burgasser (UCSD), Robert A. Simcoe (MIT), John J. Bochanski (MIT), Didier Saumon (LANL), Eric E. Mamajek (U. Rochester), Michael C. Cushing (NASA/JPL), Mark S. Marley (NASA/Ames), Craig McMurtry (U. Rochester), Judith L. Pipher (U. Rochester) and William J. Forrest (U. Rochester)

Three-color image of the sky around TWA 30, with the faint red companion TWA 30B indicated due north.

Earlier this year, U. Hawaii Graduate student Dagny Looper reported the discovery of a young, low-mass and highly active star in the TW Hydrae Association, TWA 30.  She has now found that that star is just one component of an unusual pair.

Only 80″ (0.02 degrees) north of TWA 30 lies a faint red star just barely visible in the image above.  This source is so red that it was worth a look with the MagE spectrograph at the Magellan Telescopes.  The resulting spectrum revealed a rich set of the same forbidden emission lines the we saw in the spectrum of TWA 30, indicating accretion onto the star from a circumstellar disk and  powerful magnetically-driven jets emanating from the star’s poles.  Moreover, we found two lines of forbidden neutral carbon emission, [C I], an important probe of the gas in the disk that surrounds this source, according to a 2009 study led by Dr. Barbara Ercolano.  This is the first time these lines have been seen in the spectrum of a young star.

Two spectra (black and red lines) of TWA 30B taken with the MagE spectrograph, highlighting the bevy of emission lines arising from accretion onto and jets powered by this young star. The forbidden carbon lines are seen in the bottom panel around 8700 A and 9800 A

Like TWA 30, the new source is highly variable in the near-infrared, changing its color by over a  magnitude over the course of a few days.  However, its “direction” of variability on a near-infrared color-color plot is different from that of TWA 30, indicating that it is caused not by changes of line-of-sight reddening, but changes in how its disk is blocking starlight.  The disk, apparently viewed edge-on, has structure to it!

Variations in the near-infrared colors of TWA 30A (red stars) and TWA 30B (green stars). TWA 30A's changes are consistent with reddening caused by dust extinction (arrow at lower right), but TWA 30B's variations appear to be due to patchy obscuration by a circumstellar disk.

The motion of the new source both across the sky and radially away from us, is consistent with being both a member of the TW Hydrae association and a co-moving companion to TWA 30.  Hence, these two stars – TWA 30A and TWA 30B – are a remarkable wide young binary system, nearly 3400 AU in projected separation.  What is most surprising is that TWA 30B is about 5 magnitudes fainter than TWA 30A in the near-infrared, even though it has an earlier spectral type.  This is further evidence that the star is being obscured by an edge-on disk, much like the faint secondary in the HK Tau B binary star system.  Upcoming Hubble Space Telescope observations led by Dr. John Bochanski should reveal more about this remarkable system.

This result was published in in the Astronomical Journal. Authors include Dagny L. Looper (U. Hawaii), John J. Bochanski (MIT/Penn. State), Adam J. Burgasser (UCSD), Subhanjoy Mohanty (Imperial College), Eric E. Mamajek (U. Rochester), Jacqueline K. Faherty (SUNY Stony Brook), Andrew A. West (Boston U.), and Mark A. Pitts (U. Hawaii).

Two for the price of one

Spitzer Science Center colleague Christopher Gelino and I report the identification of low-mass binary system that refuses to show itself.  The source, 2MASS J20261584–2943124, is an L dwarf which until now had seemed to be a perfectly unassuming source.  However,  low-resolution, near-infrared spectroscopy we obtained with the IRTF SpeX spectrograph revealed a peculiar absorption feature that is commonly seen in very low-mass “spectral binaries”, blends of stars with different spectral types.  Our analysis indicates that this source is an L dwarf plus T dwarf pair, with a relative brightness of roughly 4 magnitudes (or 40 times fainter) in the near-infrared.  We were unable to resolve the putative pair with the Keck Observatory laser guide star adaptive optics system, which rules a binary wider than  9 Astronomical Units (about the distance between the Sun and Saturn).  Next step: look for Doppler shifts in the spectrum that would indicate the gravitational influence of the unseen companion and allow us to measure its mass.

This research was published in the Astronomical Journal. Authors include Christopher R. Gelino (Spitzer Science Center) and Adam J. Burgasser (UCSD).

July 2010

Not all brown dwarfs are brown

We report observations of an unusually blue brown dwarf, a nearby object that may be among the coldest and oldest brown dwarfs known.  The source, ULAS J141623.94+134836.3, was originally discovered in the UKIDSS survey independently by R. Scholz and B. Burningham et al., and early results indicated its surface could be as cool as 500 K. It could even be the first Y dwarf.   Our near-infrared spectrum, obtained with the IRTF SpeX spectrograph, instead shows it to be somewhat warmer (650 K), as well as old, massive and depleted in “metals” (any element other than hydrogen and helium).  ULAS J1416+1348 is also a companion to the unusually blue L dwarf SDSS J141624.08+134826.7 discovered earlier this year by several groups.  This nearby brown dwarf pair has generated a lot of interest among astronomers, with five publications in six months.

This result was published in in the Astronomical Journal; it was also an IRTF science highlight.

June 2010