arXiv:1009.5722v1  [astro-ph.SR]  28 Sep 2010
Submitted to ApJ 16 Aug 2010; Accepted for publication 28 Sep 2010
Preprint typeset using LATEX style emulateapj v. 11/10/09
CLOUDS IN THE COLDEST BROWN DWARFS: FIRE SPECTROSCOPY OF ROSS 458C
Adam J. Burgasser1,2,3, Robert A. Simcoe2,4, John J. Bochanski2, Didier Saumon5, Eric E. Mamajek6, Michael C.
Cushing7, Mark S. Marley8, Craig McMurtry6, Judith L. Pipher6, and William J. Forrest6
Submitted to ApJ 16 Aug 2010; Accepted for publication 28 Sep 2010
ABSTRACT
Condensate clouds are a salient feature of L dwarf atmospheres, but have been assumed to play
little role in shaping the spectra of the coldest T-type brown dwarfs. Here we report evidence of
condensate opacity in the near-infrared spectrum of the brown dwarf candidate Ross 458C, obtained
with the Folded-Port Infrared Echellette (FIRE) spectrograph at the Magellan Telescopes. These data
verify the low-temperature nature of this source, indicating a T8 spectral classification, log10 Lbol/L?
= ?5.62?0.03, Teff = 650?25 K, and a mass at or below the deuterium burning limit. The data
also reveal enhanced emission at K-band associated with youth (low surface gravity) and supersolar
metallicity, reflecting the properties of the Ross 458 system (age = 150?800 Myr, [Fe/H] = +0.2 to
+0.3). We present fits of FIRE data for Ross 458C, the T9 dwarf ULAS J133553.45+113005.2,and the
blue T7.5 dwarf SDSS J141624.08+134826.7B, to cloudless and cloudy spectral models from Saumon
& Marley. For Ross 458C we confirm a low surface gravity and supersolar metallicity, while the
temperature differs depending on the presence (635+25?35 K) or absence (760+70?45 K) of cloud extinction.
ULAS J1335+1130and SDSS J1416+1348Bhavesimilartemperatures(595+25?45 K), but distinct surface
gravities (logg = 4.0?4.5 cgs versus 5.0?5.5 cgs) and metallicities ([M/H] ? +0.2 versus -0.2). In all
three cases, cloudy models provide better fits to the spectral data, significantly so for Ross 458C.
These results indicate that clouds are an important opacity source in the spectra of young cold T
dwarfs, and should be considered when characterizing the spectra of planetary-mass objects in young
clusters and directly-imaged exoplanets. The characteristics of Ross 458C suggest it could itself be
regarded as a planet, albeit one whose cosmogony does not conform with current planet formation
theories.
Subject headings: stars: fundamental parameters ? stars: individual (Ross 458C
(ULAS J130041.72+122114.7), ULAS J133553.45+113005.2,
SDSS J141624.08+134826.7B (ULAS J141623.94+134836.3)) ? stars: low
mass, brown dwarfs ? stars: planetary systems
1. INTRODUCTION
Mineral condensate clouds are a unique and promi-
nent constituent of the atmospheres of the coldest
brown dwarfs, the L dwarfs and T dwarfs (Kirkpatrick
2005 and references therein). For the L dwarfs, con-
densates are a primary contributor to shaping spec-
tral energy distributions (SEDs). Their formation de-
pletes the atmosphere of TiO and VO gases, trans-
forming optical SEDs and defining the L dwarf spec-
tral sequence (Jones & Tsuji 1997; Kirkpatrick et al.
1999; Lodders 2002). Condensate opacity contributes
to muted infrared H2O bands and the characteris-
tic red near-infrared colors of L dwarfs (J ? K =
1 Center for Astrophysics and Space Science, University
of California San Diego, La Jolla, CA 92093, USA; abur-
gasser@ucsd.edu
2 Massachusetts Institute of Technology, Kavli Institute for
Astrophysics and Space Research, Building 37, Room 664B, 77
Massachusetts Avenue, Cambridge, MA 02139, USA
3 Hellman Fellow
4 Alfred P. Sloan Foundation Research Fellow
5 Los Alamos National Laboratory, P.O. Box 1663, MS F663,
Los Alamos, NM 87545, USA
6 Dept. of Physics and Astronomy, University of Rochester,
Rochester, NY USA 14627
7 Jet Propulsion Laboratory, California Institute of Technol-
ogy, Pasadena, CA, USA
8 NASA Ames Research Center, Mail Stop 245-3, Moffett
Field, CA 94035, USA
1.5...2.5; Chabrier et al. 2000; Knapp et al. 2004). Ev-
idence of silicate grain absorption has also been found
in the mid-infrared spectra of L dwarfs (Cushing et al.
2006; Helling et al. 2006; Looper et al. 2008). For the
T dwarfs, condensates appear to play a minor role,
as evidenced by strong molecular gas absorption fea-
tures and typically blue near-infrared colors (J ? K =
?0.5...0.5; Burgasser et al. 2002a, 2006b; Geballe et al.
2002; Cushing et al. 2005). This shift is believed to
be the result of clouds sinking below the visible pho-
tosphere (Ackerman & Marley 2001; Cooper et al. 2003;
Tsuji 2005) or breaking apart (Burgasser et al. 2002b;
Marley et al. submitted), although the details of this
transition remain poorly understood (Burrows et al.
2006; Burgasser 2007; Saumon & Marley 2008).
Even without considering condensates, modeling the
highly structured SEDs of T dwarfs remains a chal-
lenge. The abundances and opacities of prominent
gas species ? H2 (collision induced absorption; Linsky
1969), H2O, CH4, NH3 and neutral alkalis (pressure-
broadened wings; Burrows & Volobuyev 2003) ? are
highly sensitive to photospheric gas conditions. Tem-
perature and pressure variations, elemental abundances,
and vertical mixing all have observable effects on T
dwarf spectra (e.g., Knapp et al. 2004; Burgasser et al.
2006a; Saumon et al. 2007; Stephens et al. 2009). In
principle, one could disentangle the contributions of
2 Burgasser et al.
these atmospheric properties on observed spectra and
derive bulk characteristics?mass, age and metallicity?
for individual sources (e.g., Burgasser et al. 2006a;
Leggett et al. 2009). However, a variety of effects
drive systematic deviations between model and ob-
served spectra. These include incomplete opacities
for key absorbers such as CH4, NH3 and K I (e.g.,
Allard et al. 2003; Cushing et al. 2008; Freedman et al.
2008); non-equilibrium chemistry arising from vertical
mixing (e.g., Saumon et al. 2006, 2007); metallicity ef-
fects on chemical pathways (e.g., Lodders & Fegley 2002;
Burgasser et al. 2003; Visscher et al. 2010); and the in-
fluence of condensate clouds in regions of minimum
gas opacity (e.g., Tsuji et al. 1999; Ackerman & Marley
2001; Burrows et al. 2002). These processes lead to per-
sistent uncertainties in inferred atmospheric and physical
parameters for individual sources (cf., Burningham et al.
2010; Burgasser et al. 2010b).
Benchmark T dwarfs, with independent determina-
tions of age, mass and metallicity, are important calibra-
tors for brown dwarf atmospheric studies (Pinfield et al.
2006). Such benchmarks include widely-separated
T dwarf companions to nearby stars, several exam-
ples of which have been identified over the past 15
years (e.g., Nakajima et al. 1995; Burgasser et al. 2000;
Luhman et al. 2007; Burningham et al. 2009). Most re-
cently, Goldman et al. (2010) and Scholz (2010a) re-
ported the discovery of a faint (J = 16.67?0.01),
co-moving candidate T dwarf companion to the Ross
458 system in the UKIRT Infrared Deep Sky Sur-
vey (UKIDSS; Lawrence et al. 2007). The source,
ULAS J130041.72+122114.7 (hereafter Ross 458C), is
separated by 102?? from the M0.5 Ve + M7 Ve Ross
458AB pair, implying a projected separation of over
1100 AU at a distance of 11.7?0.2 pc (van Leeuwen
2007). The late-type nature of Ross 458C was in-
ferred from its optical and near-infrared colors (z ?J =
3.55?0.19; J ?K = -0.21?0.06), faint absolute magni-
tude (MJ = 16.40?0.04)and large flux contrast in filters
sampling the 1.6 ?m CH4 band (cf., Tinney et al. 2005).
From these photometric measurements, Scholz (2010a)
and Goldman et al. (2010) estimated spectral types of
T7?1 and T8-9 for Ross 458C, respectively. However,
no spectrum was reported in either study.
If Ross 458C is a brown dwarf, it is a potentially
interesting benchmark for several reasons. Kinemat-
ics of the Ross 458 system coincide with the Hyades
supercluster (Eggen 1960; Montes et al. 2001), an ex-
tended population of relatively young and metal-rich
stars that includes the 625?50 Myr Hyades open clus-
ter (Perryman et al. 1998). The primary exhibits con-
spicuous activity in H? and Ca II lines (Browning et al.
2010), and is an X-ray source (Kiraga & Stepien 2007;
L?opez-Santiago et al. 2009). Ross 458C could there-
fore be a young and/or metal-rich brown dwarf bench-
mark, useful for constraining both evolutionary and
atmospheric models (cf. Zapatero Osorio et al. 2004;
Dupuy et al. 2009a).
In this article, we present the first spectroscopic obser-
vations of Ross 458C, obtained in the near-infrared with
the newly-commissioned Folded Port Infrared Echellette
(FIRE; Simcoe et al. 2008). These data confirm the late-
type T dwarf nature of the source, and indicate a low
surface gravity and supersolar metallicity consistent with
characteristics of the primary. Furthermore, model fits
indicate that cloud opacity plays a prominent role in
shaping its near-infrared spectrum. In Section 2 we con-
strain the age and metallicity of the Ross 458 system
based on the photometric, spectral, kinematic and rota-
tional properties of its M0.5 Ve primary. In Section 3 we
describe our FIRE observations of Ross 458C and two
other late-type T dwarfs, from which we derive spectral
types and basic physical characteristics. In Section 4 we
fit the spectral data for these three sources to cloudless
and cloudy atmospheric models from Saumon & Marley
(2008, hereafter SM08), and compare the inferred atmo-
spheric and physical parameters to prior studies. In Sec-
tion 5 we discuss the role of condensates in shaping the
near-infrared SEDs of young, cold T dwarfs, and their
influence on T dwarf temperature scales. In Section 6
we summarize our results.
2. CHARACTERIZING THE ROSS 458 SYSTEM
2.1. Examining Association with the Hyades Open
Cluster
Under the assumption of coevality, some physical prop-
erties of Ross 458C can be inferred from its stellar com-
panions, specifically age and metallicity. As such, it is
necessary to examine the association of this system with
the Hyades supercluster, a kinematic stream composed
of several dozen nearby stars with similar space motions
(Eggen 1960; Montes et al. 2001). While co-moving as-
sociations typically have similar ages and compositions,
the ages of the Hyades supercluster stars span the star-
formation history of the disk (0.4?2.0 Gyr; Famaey et al.
2005; Antoja et al. 2008), so stream membership is not
a particularly useful age indicator. Within the stream
is the eponymous Hyades open cluster, the nearest open
cluster to the Sun, whichi constitutes a coeval group with
an age of 625?50 Myr (Perryman et al. 1998). Because
of its proximity, members of the Hyades cluster halo are
distributed broadly across the sky. It is therefore not un-
reasonable for Ross 458 to be associated with this struc-
ture.
To examine this possibility, we calculated the systemic
spacevelocity9 ofthe Ross 458system. We used the long-
baseline proper motion from Tycho-2 (?? = ?640.1?1.5
mas yr?1; ?? = ?25.1?1.4 mas yr?1; H?g et al. 2000),
the radial velocity from Nidever et al. (2002, Vr = -
11.2?0.1 km s?1), and the revised Hipparcos parallax
(pi = 85.5?1.5 mas; van Leeuwen 2007). From these we
find heliocentric(U, V, W)= (30.6, -18.8, -9.9)?(0.5, 0.3,
0.2) km s?1. The velocity of the Hyades cluster in the
same coordinate system (de Bruijne et al. 2001) is (U, V,
W) = (42.25, -19.06, -1.45)?(0.14, 0.13, 0.21) km s?1, a
difference of (11.7, -0.3, 8.5)?(0.5, 0.3, 0.3) km s?1 (see
also Perryman et al. 1998). This is much larger than
the 1D velocity dispersion of Hyades cluster members,
?0.35 km s?1 (Madsen 2003), amounting to a ?20? dis-
crepancy. Hence, Ross 458 is not currently kinematically
associated with the Hyades open cluster.
Has Ross 458 ever been associated with the Hyades
9 Weadopted aheliocentric, right-handed rectangular coordinate
system with U velocity locally pointing away from the Galactic
center, V velocity pointing in the direction of Galactic rotation (?,
b = 90?, 0?) and W velocity pointing towards the north Galactic
pole (b = +90?).
Clouds in Ross 458C 3
cluster? Its current separation from the cluster center is
53 pc. Using the epicyclic orbit approximation to model
the past orbits of Ross 458 and the Hyades (adopting
the Oort parameters of Feast & Whitelock 1997 and Lo-
cal Standard of Rest from Dehnen & Binney 1998), we
find that Ross 458 was never closer to the Hyades than
?50 pc during the past 100 Myr, or about 5 tidal radii.
Hence, there is no evidence of past or present physical as-
sociation between Ross 458 and the Hyades open cluster,
and association with the Hyades supercluster provides no
meaningful constraint on the age of this system.
2.2. The Age of the Ross 458 System
Despite the wide range of ages associated with the
Hyades kinematic stream there is considerable evidence
that the Ross 458 system itself is quite young. The
M0.5 Ve primary10 (Reid et al. 1995) is an ?ultra-
fast rotator? with a period between 1.54 and 2.89
days (Pizzolato et al. 2003; Kiraga & Stepien 2007).
Ross 458A had the highest measured projected rota-
tional velocity (vsini = 9.7?0.5 km s?1) among 69
nearby M0-M3 dwarfs in the California Planet Search
program, in which only ?3% of early-M dwarfs had any
detectable rotation (vsini greaterorsimilar 3 km s?1; Browning et al.
2010). Beuzit et al. (2004) have argued for an age less
than 1 Gyr on the basis of this rotation.
In addition, Ross 458A exhibits strong magnetic ac-
tivity. The presence of H? emission indicates an up-
per activity age limit of 400?800 Myr (West et al. 2008).
This limit is roughly consistent with the analysis of
Takeda et al. (2007), who deduce a 68% upper limit
of 440 Myr based on theoretical modeling of the pri-
mary?s high-resolution spectrum (however, this does not
appear to be a ?well-defined? age according to that
study). Ross 458 is also an X-ray source, detected in
three R?ontgen Satellite (ROSAT) Position Sensitive Pro-
portional Counter (PSPC) pointings and in the All-Sky
Survey (Voges et al. 2000). Combining these four obser-
vations yields a mean ROSAT PSPC flux in the 0.1-2.4
keVband of4.35cts?1, and mean hardnessratiosofHR1
= -0.13 and HR2 = -0.09. The mean X-ray flux, cou-
pled with the energy conversion factors of Fleming et al.
(1995) and the Hipparcos parallax, implies a soft X-ray
luminosity of LX = 5?1029 ergs?1 and an X-ray to bolo-
metric luminosity ratio of logLX/Lbol = -2.7. Such a
high ratio defines Ross 458 as a?saturated? X-ray emit-
ter, as expected for stars of similar color and rotation
periods of lessorsimilar4 days (Pizzolato et al. 2003). The combi-
nation of color and rotation period for Ross 458A also
classifies it as a C-sequence (?convective?) rotator in the
scheme of Barnes (2003). The characteristic spin-down
timescale for such stars is ?300 Myr. From a cross-
comparison of age-dated cluster and field populations of
low-mass, solar-type stars, Barnes (2003) concludes that
the fraction of C-sequence ultrafast rotators reduces to
zero by age ?800 Myr. Hence, this age can be consid-
ered a robust upper limit for the Ross 458 system that
is consistent with activity estimates.
10 The M dwarf components of Ross 458 are separated by 0.??5
(Beuzit et al. 2004), implying that reported optical spectroscopy
of this system probably includes both components. However,
Ross 458B is ?8 mag fainter at V (Section 2.3), implying that
its contribution to combined light optical spectra is negligible.
6000 6500 7000 7500 8000
Wavelength (Angstroms)
0.0
0.5
1.0
1.5
Normalized F
?
M0 template
M1 template
M2 template
Ross 458AB
Telluric
Telluric
Telluric
6450 6500 6550 6600 6650 6700 6750 6800
Wavelength (Angstroms)
0.5
0.6
0.7
0.8
0.9
1.0
1.1   H?
  Li I
Figure 1. Red optical spectrum of Ross 458AB (black line) com-
pared to M0 (red line), M1 (green line) and M2 (blue line) spectral
templates from Bochanski et al. (2007). All spectra are normal-
ized at 7500 ?A, and templates are smoothed to the resolution of
the observational data (?/?? ? 800) using a gaussian kernel. Re-
gions of strong telluric absorption in the observational data are
indicated. Inset box details the 6450?6800 ?A region highlighting
the prominent H? emission and absence of Li I absorption.
A lower limit on the age of this system can be in-
ferred from the optical spectrum of the primary. Fig-
ure 1 displays ?/?? ? 800 optical spectroscopy for the
combined AB components obtained by J. D. Kirkpatrick
with the CTIO-1.5mRCSpectrographon22May199611.
The spectrum is compared to M0, M1 and M2 spectral
templates from Bochanski et al. (2007), the M1 show-
ing the strongest agreement with the data. The promi-
nent 6563 ?A H? line is clearly seen in these data, but
there is no indication of Li I absorption at 6708 ?A, im-
plying an ageolder than 30-50Myr (Magazzu et al. 1993;
Chabrier et al. 1996; Bildsten et al. 1997). Furthermore,
there is no indication low surface gravity features, such
as the weakened 7700 ?A K I doublet lines observed of
cooler young M dwarfs up to the age of the Pleiades
(e.g., Martin et al. 1996; Kirkpatrick et al. 2008). These
spectroscopic constraints place a lower limit on the age
of this system of ?150 Myr.
2.3. The Metallicity of the Ross 458 System
The metallicity of Ross 458A can be determined
from its position on the V ? K/MK color magni-
tude diagram (CMD), a combination that effectively
segregates mass and metallicity variations in field
M dwarfs (Bonfils et al. 2005; Johnson & Apps 2009;
Schlaufman & Laughlin 2010). For better accuracy, we
disentangled the flux of Ross 458A from its closely-
separated M7 companion. Using ?K = 4.41 from
Beuzit et al. (2004, a 0.1 mag uncertainty is assumed),
wedeterminedMKA = 5.26?0.04andMKB = 9.67?0.11.
We then used unresolved Hipparcos/2MASS photome-
try (V ? K = 4.180?0.016) and the mean V ? K/MK
relation of Schlaufman & Laughlin (2010) to estimate
?V = 7.93?0.14 and hence (V ? K)A = 4.16?0.02.
The MK/metallicity calibration of Johnson & Apps
(2009) then yields [Fe/H] = +0.31?0.05, while the
11 These data are available at DwarfArchives,
http://dwarfarchives.org.
4 Burgasser et al.
V ?K/metallicity calibration of Schlaufman & Laughlin
(2010) yields [Fe/H] = +0.20?0.05. Hence, both rela-
tions indicate a supersolar metallicity for the Ross 458
system. However, these quantitative estimates should
be regarded with care given the young age of the system.
The primary may be systematically brighter/redderthan
the control samples used by Johnson & Apps (2009) and
Schlaufman & Laughlin (2010), resulting in a systemat-
ically higher metallicity.
In summary, the Ross 458 system has a probable age
of 150?800 Myr based on its rotation, strong activity
indicators and spectroscopic features. It also appears to
be metal-rich, with [Fe/H] = +0.2 to +0.3, albeit with
possible age biases. We will compare these values to
results from spectral modeling of the T dwarf companion
in Section 4.
3. NEAR-INFRARED SPECTROSCOPY
3.1. The FIRE spectrograph
FIRE is a cross-dispersed, near-infrared spectrometer
recently installed at the auxiliary Nasmyth focus of the
Magellan 6.5-meter Baade Telescope. It is a single-object
spectrograph, designed primarily to deliver moderate-
resolution (?/?? ? 6000 for a 0.??6-wide slit) spectra
over the 0.85?2.5 ?m range in a single exposure. The 21
orders dispersed by a combination of prisms and reflec-
tion gratingareimagedonto a2048?2048,HAWAII-2RG
(H2RG) focal plane array. A mirror can also be rotated
in place of the reflection grating to obtain a single-order,
high-throughput spectrum at lower resolution (?/?? =
250-350; see below). A second H2RG camera images the
entrance slit through an MKO12 J-band filter for source
acquisition and manual guiding. Further details on this
instrument can be found in Simcoe et al. (2008, 2010)
and on the instrument webpage.13
3.2. Observations and Data Reduction
Data presented here were obtained during FIRE?s com-
missioning run during 31 March through 7 April 2010
(UT). Observations were obtained for three T dwarfs:
Ross 458C, the T7.5 ULAS J141623.94+134836.3
(aka SDSS J141624.08+134826.7B and here-
after SDSS J1416+1348B; Burningham et al.
2010; Scholz 2010b; Burgasser et al. 2010b) and
the T9 ULAS J133553.45+113005.2 (hereafter
ULAS J1335+1130; Burningham et al. 2008). A
summary of the observations is given in Table 1.
Conditions ranged from clear (3 and 7 April) to patchy
cirrus (6 April), with seeing of 0.??6?0.??8 at J-band. For
all sources, we employed FIRE?s prism-dispersed mode
with the 0??6 wide slit aligned to the parallactic angle
(rotator angle 89.?5). The resolution for this setting
ranges from ?/?? ? 350 in the 1.0?1.4 ?m Y- and
J-band region to ?/?? ? 250 in the 1.4?2.5 ?m H-
and K-band region, as determined from gaussian fits
to emission lines in the arc lamp spectra and telluric
emission lines in the science images. The spectrograph
detector was read out using the 4-amplifier mode at
?high gain? (1.2 counts/e?). Each source was observed
12 Mauna Kea Observatory filter system; see Tokunaga et al.
(2002) and Simons & Tokunaga (2002).
13 http://www.firespectrograph.org
in four 150 s exposures using an ABBA dither pattern,
nodding 9?? along the 30?? slit, followed by observation
of a nearby A0 V star at similar airmass for telluric
correction and flux calibration. Exposures of arc lamps
(HeNeAr on 3 April; NeAr on 6 and 7 April) and quartz
lamps, reflected off of the Baade secondary screen, were
obtained with the target and A0 V stellar observations
for wavelength and pixel response calibration.
Data were reduced using a combination of standard
IRAF14 and custom IDL15 routines. A pixel response
frame was first constructed by combining two quartz
lamp images taken with the lamp at high (2.5 V) and
low (1.5 V) voltage settings. This was necessary to com-
pensate for the substantial thermal emission from the
lamp screen which tended to saturate the detector at the
longest wavelengths. The two images were stitched to-
gether at column 1300(roughly 1.4?m), then normalized
by a median combination of all columns along the disper-
sion region, which eliminated the intensity step function
at the stitch and isolated pixel-to-pixel response varia-
tions. The science images were divided by this pixel re-
sponse frame, and bad pixels interpolated by averaging
over nearest neighbors. The images were then pairwise-
subtracted to remove sky background. Spectra were ex-
tracted using the IRAF apall task. Calibrator stars were
extracted first, then used as traces for the T dwarf and
arclamp spectra. A wavelength solution was generated
from a fifth-order spline fit to 60 lines in the arclamp
spectra using the IRAF task identify, producing a mean
dispersion of 8.5 ?A pixel?1 and scatter of 9 ?A (?1 pixel).
Note that the actual dispersion varies considerably along
the chip, from 4 ?A pixel?1 at 1.0 ?m to 20 ?A pixel?1 at
2.4 ?m. The individual spectra were combined using the
xcombspec routine from SpeXtool (Cushing et al. 2004),
using a robust weighted mean after normalizing in the
1.2?1.4 ?m region. The SpeXtool task xtellcor general
was used to compute telluric correction and flux calibra-
tion from the A0 V star, following the prescription of
Vacca et al. (2003) and assuming a line profile width of
18 ?A.
To verify the flux calibration of our data, we com-
puted MKO spectrophotometric colors from all three
sources from the FIRE spectra following Cushing et al.
(2005), computing 100iterationswith each spectrum var-
ied by its noise spectrum to determine uncertainties.
For ULAS J1335+1130 and SDSS J1416+1348B, these
colors match UKIDSS photometry to within the uncer-
tainties, indicating robust relative flux calibration. For
Ross 458C, the MKO colors are intermediate between
the UKIDSS and ground-based GROND (Greiner et al.
2008) photometry reported in Goldman et al. (2010),
which themselves differ significantly (UKIDSS J ? K
= 0.51?0.06; GROND J ? K = -0.21?0.06). This
discrepancy may be due to differences in the filter
basspands between UKIDSS, MKO and GROND fil-
ter passbands, or may reflect intrinsic variability in the
source. In Figure 2, we compare our reduced FIRE
14 Image Reduction and Analysis Facility (IRAF; Tody 1986) is
distributed by the National Optical Astronomy Observatory, which
is operated by the Association of Universities for Research in As-
tronomy, Inc., under cooperative agreement with the National Sci-
ence Foundation.
15 Interactive Data Language.
Clouds in Ross 458C 5
0.0
0.2
0.4
0.6
0.8
1.0
1.2
Normalized 
f ?
ULAS J1335+1130
FIRE
ISAAC/NIRI
1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4
Wavelength (?m)
-0.2-0.1
0.00.1
0.2
FIRE-NIRI
Figure 2. Spectral observations of ULAS J1335+1130 based on Magellan/FIRE (this study) and Gemini-North/NIRI plus VLT/ISAAC
(Burningham et al. 2008, red line). Both spectra are normalized at the 1.27 ?m flux peak. The difference between these spectra (bottom
panel) is less than 10% over most of the near-infrared band.
Table 1
Magellan/FIRE Observing Log.
Source SpT UKIDSS J Date (UT) Airmass tinta A0 V Calibrator
Ross 458C T8 16.69?0.01 3 Apr 2010 1.34?1.36 642 HD 114381
ULAS J133553.45+113005.2 T9 17.90?0.01 7 Apr 2010 1.38?1.42 642 HD 114381
ULAS J141623.94+134836.3b T7.5p 17.35?0.02 6 Apr 2010 1.40?1.44 642 HD 124773
a Total integration time divided into four ABBA dithered pointings. Total integration includes
10.61 s frame readout.
b aka SDSS J141624.08+134826.7B.
spectrum of ULAS J1335+1130 to that obtained by
Burningham et al. (2008) using the Gemini-North Near
Infrared Imager and Spectrometer (NIRI; Hodapp et al.
2003) and Paranal Very Large Telescope Infrared Spec-
trometer and Array Camera (ISAAC; Moorwood et al.
1998). The Burningham et al. spectrum was stitched to-
gether from four orders individually flux-calibrated using
UKIDSS YJHK photometry. Our spectrum is in excel-
lent agreement with the NIRI and ISAAC data, with de-
viations of less than 10% outside the telluric absorption
regions. Importantly, the FIRE data reproduce the rel-
ative YJHK flux peaks, which are essential for spectral
model fits (Section 4).
3.3. Results
Figure 3 displays the reduced spectrum of Ross 458C,
scaled to its absolute flux (F? units at 10 pc) based on
UKIDSS J-band photometry and the Hipparcos parallax
of the Ross 458 system. Signal-to-noise (S/N) in the
spectrum ranges from 50 to 80 in each of the YJHK flux
peaks. Ross 458C exhibits the classic spectral signatures
ofa T dwarf, with strong H2O and CH4 absorptionbands
at 1.15, 1.4, 1.6, 1.8 and 2.2 ?m; a steep red optical slope
shaped by the pressure-broadened 0.77 ?m K I wings;
anda relativelyblue spectralenergydistribution from 1.2
to 2.1 ?m, due in part to collision-induced H2 absorption
(Linsky 1969; Saumon et al. 1994; Borysow 2002). As
anticipated from its extreme CH4-filter colors, the 1.6?m
6 Burgasser et al.
1.0 1.5 2.0
Wavelength (?m)
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
F ?
 (mJy)
Ross 458C
H2O H2O
H2O
CH4
CH4
CH4
CH4
K I
CIA H2
1.23 1.24 1.25 1.26
0.85
0.90
0.95
1.00
1.05
1.10
F ?
 / Continuum
K I
Figure 3. FIRE prism spectrum of Ross 458C in absolute flux units (F?). The uncertainty spectrum is indicated by the grey line.
Prominent H2O and CH4 absorption features are labeled, as well as regions influenced by the pressure-broadened K I doublet wing
(? lessorsimilar 1 ?m) and collision-induced H2 opacity (? greaterorsimilar 1.75 ?m). The inset box shows continuum-normalized flux between 1.23 and 1.26 ?m,
highlighting the weak 1.24/1.25 ?m K I doublet present in the spectrum of this source (red lines show gaussian fits).
CH4 band of Ross 458C is quite deep, while the J- and
H-band flux peaks are narrow, all indicative of a very
late-type, low-temperature T dwarf. Indeed, as shown in
Figure 4, these features in the spectrum of Ross 458C are
intermediate between those of SDSS J1416+1348B and
ULAS J1335+1130, indicating an intermediate spectral
type.
Our spectrum of Ross 458C also reveals the pres-
ence of the 1.24/1.25 ?m K I doublet, highlighted in
the inset box in Figure 3. These lines are known to
fade in the latest-type T dwarfs (Burgasser et al. 2002a;
McLean et al. 2003), likely a consequence of KCl(g)
formation at temperatures below 1000 K (Lodders
1999). We measured pseudoequivalent widths pEW =
2.7?0.2 ?A for each doublet line, comparable to measure-
ments for other T7?T8 dwarfs (Burgasser et al. 2002a;
McLean et al. 2003). We do not see these lines in our
spectra of ULAS J1335+1130 or SDSS J1416+1348B
to limits of 0.5 and 0.3 ?A, respectively. For the lat-
ter source, this may be a consequence of its subso-
lar metallicity in analogy to the metal poor T6 dwarf
2MASS J09373487+2931409 (Burgasser et al. 2002a;
McLean et al. 2007; see Section 4.2)
Another feature of note in the spectrum of Ross 458C
is its fairly pronounced 2.1 ?m K-band peak. The rel-
ative brightness of this peak is greater than those of
both ULAS J1335+1130 and SDSS J1416+1348B, in ac-
cord with its reddest J ? K color. The K-band peak
is normally suppressed by H2 absorption in late-type T
dwarfs. This absorption is weakened in brown dwarfs
with low surface gravities (i.e., young, low-mass) and/or
higher metallicities, as both effects result in a lower pho-
tospheric pressure and weaker collision-induced absorp-
tion. For Ross 458C, both of these properties likely con-
tribute based on the characteristics of the system (Sec-
tion 2).
3.4. Spectral Classification
To classify the T dwarfs, we used the near-infrared
spectral indices and index-spectral type relations de-
fined in Burgasser et al. (2006b); Warren et al. (2007);
and Burningham et al. (2008); these are listed in Ta-
ble 2. Our index measurements for ULAS J1335+1130
and SDSS J1416+1348B are in rough agreement
with previous results (Burningham et al. 2008, 2010;
Burgasser et al. 2010b), with minor variances at-
tributable to differences in spectral resolution and S/N
between data sets. We derive mean classifications of
T8, T8.5 and T7.5p for Ross 458C, ULAS J1335+1130
and SDSS J1416+1348B, respectively. Uncertainties of
0.5 subtypes are assumed for the first two sources, while
the last classification is deemed ?peculiar? due to the
larger scatter in indices (?1.2 subtypes). The last two
Clouds in Ross 458C 7
1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4
Wavelength (?m)
0.0
0.5
1.0
1.5
2.0
Normalized 
F ?
 + Constant
Ross 458C
ULAS J1335+1130
SDSS 1416+1348B
Figure 4. FIRE prism spectra of (top to bottom) Ross 458C, ULAS J1335+1130 and SDSS J1416+1348B. Data are normalized at 1.27 ?m
and offset by constants (dotted lines). Note the variations in the Y- and K-band flux peaks.
Table 2
Spectral Indices for Observed T Dwarfs.
Ross 458C ULAS J1335+1130 SDSS J1416+1348B
Index Valuea SpT Valuea SpT Valuea SpT Ref.
H2O-J 0.051?0.002 T8.0 0.028?0.003 T8.5 0.044?0.003 T8.0 1
CH4-J 0.154?0.001 T8.5 0.109?0.002 T9.0 0.262?0.002 T7.0 1
WJ 0.272?0.002 T9.0 0.253?0.002 T9.0 0.360?0.002 T7.0 2,3
H2O-H 0.193?0.003 T7.5 0.192?0.007 T7.5 0.148?0.004 T9.0 1
CH4-H 0.112?0.003 T8.0 0.068?0.005 T9.0 0.192?0.003 T7.0 1
NH3-H 0.632?0.004 ??? 0.537?0.007 ??? 0.610?0.005 ??? 4
CH4-K 0.078?0.005 T7.5 0.073?0.013 T7.5 0.191?0.032 T5.5 1
K/J 0.167?0.001 ??? 0.130?0.002 ??? 0.044?0.001 ??? 1
J ?Hb -0.14?0.01 ??? -0.34?0.02 ??? -0.25?0.01 ???
H ?Kb 0.38?0.02 ??? 0.06?0.05 ??? -1.02?0.09 ???
J ?Kb 0.24?0.02 ??? -0.28?0.04 ??? -1.25?0.09 ???
References. ? (1) Burgasser et al. (2006b); (2) Warren et al. (2007); (3)
Burningham et al. (2008); (4) Delorme et al. (2008).
a Values for each source were measured for 1000 realizations of the spectrum (100
realizations for spectrophotometric colors), each with a normal distribution of ran-
dom values scaled by the noise spectrum added to the original fluxes. The reported
values are the means and standard deviations of these measurements.
b Spectrophotometric colors computed according to Cushing et al. (2005).
8 Burgasser et al.
classifications formally agree with previously published
results (Burningham et al. 2008, 2010; Burgasser et al.
2010b). Our classification of Ross 458C splits the dif-
ference between the photometric estimates of Scholz
(2010a) and Goldman et al. (2010), and is consistent
with the relative ordering of the spectra in Figure 4. The
absolute J magnitude of Ross 458C is also comparable
to other T8 dwarfs (Vrba et al. 2004).
3.5. Physical Properties of Ross 458C
With a spectral type and known distance, it is pos-
sible to assess some of the basic physical properties of
Ross 458C. A bolometric luminosity was determined
using the absolute UKIDSS JHK-band magnitudes
and bolometric corrections (BC) from Liu et al. (2010).
Adopting16 BCJ = 2.62?0.14, BCH = 2.19?0.07 and
BCK = 2.05?0.09 for a T8 dwarf, we derive estimates of
log10 Lbol/L? = -5.69?0.06, log10 Lbol/L? = -5.65?0.04
and log10 Lbol/L? = -5.55?0.05 based on J-, H- and
K-band fluxes, or a weighted mean of log10 Lbol/L? =
-5.62?0.03. This luminosity is similar to that of the
T8 spectral standard 2MASS J04151954-0935066 (here-
after 2MASS J0415-0935; log10 Lbol/L? = -5.67?0.02;
Saumon et al. 2007). The Teff of Ross 458C can be
inferred from its luminosity by adopting a radius from
the evolutionary models of SM08. For an age of 150?
800 Myr (Section 2.2), these models predict a radius of
0.12?0.13 R?, Teff = 650?25 K and a mass in the range
0.006?0.011 M?. The inferred temperature is cooler
than that of 2MASS J0415-0935 (Teff = 750?25 K;
Saumon et al. 2007) due to the inflated radius, while
the inferred mass is below the deuterium burning min-
imum mass limit (0.014 M?; Chabrier & Baraffe 2000;
Spiegel et al. 2010).
4. SPECTRAL MODEL FITS
4.1. Spectral Models and Method
We compared the FIRE spectra of these three
T dwarfs to a suite of atmosphere models from
Marley et al. (2002) and SM08, following the prescrip-
tions detailed in Cushing et al. (2008); Burgasser et al.
(2008b) and Bowler et al. (2009). The one-dimensional
models include all significant gas species consid-
ered in Freedman et al. (2008) as well as CO2 (cf.
Yamamura et al. 2010), albeit with line lists for CH4
and NH3 that are known to be incomplete (e.g.,
Saumon et al. 2006). Solar abundances from Lodders
(2003) were assumed. Non-equilibrium chemistry of CO,
CH4, H2O and NH3 (the relevant molecules influencing
the near-infrared spectra of T dwarfs) by vertical trans-
port was accounted for using an eddy diffusion parame-
terization, Kzz (Fegley & Lodders 1996; Griffith & Yelle
1999; Saumon et al. 2006; Hubeny & Burrows 2007).
Equilibrium condensate formation was included in
the chemistry17, with grain size and vertical distri-
butions parameterized by a sedimentation rate fsed
16 Error estimates quoted here take into account uncertainties in
the spectral classification (?0.5 subtype), photometry, parallax of
Ross 458 and the spectral type/BC relations (0.14, 0.07 and 0.08
mag for J, H and K, respectively). Systematic errors?e.g., from
gravity or metallicity effects?are not accounted for.
17 The chemical equilibrium calculation accounts for all im-
portant condensed species (Lodders & Fegley 2006; Visscher et al.
2010). For the cloud opacity we included only silicates, iron, and
(Ackerman & Marley 2001; Marley et al. 2002). While
contemporaneous studies of late-type T dwarfs have
generally ignored the opacity from condensate grains
(e.g., Leggett et al. 2007, 2009; Saumon et al. 2007;
Burgasser et al. 2008b), we considered two sets of SM08
models with (?cloudy?) and without (?cloudless?) con-
densate opacity included. Both sets of models use the
same gas chemistry and have the same constituent abun-
dances at any given pressure and temperature. The
cloudless models assume that condensates do not con-
tribute to the opacity (but do contribute to the gaschem-
istry), perhaps because they have ?rained out? of the at-
mosphere or occupy only a small fraction of the visible
surface area. The cloudy models explicitly include opac-
ity from cloud grains. The cloudless models sampled at-
mospheric parameters Teff = 500?1000 K (50 K steps);
logg = 4.0?5.5 cgs (0.5 dex steps); [M/H] = ?0.3, 0 and
+0.3 dex relative to solar; and Kzz = 0 and 104 cm2 s?1.
The cloudy models sampled the same parameter space
with fsed = 2 and solar metallicity alone. Hereafter, we
refer to sets of parameters by the notation k = (Teff,
logg, [M/H], fsed, logKzz), where fsed = NC for the
cloudless models. The model spectra were computed as
surface fluxes (F? flux units), smoothed to the mean res-
olution of the FIRE prism data (?/?? = 300) using a
gaussian kernel, and interpolated onto the same wave-
length scale as the data. We also assigned physical pa-
rameters of mass, age and radius to each spectral model,
interpolating the appropriate evolutionary tracks from
SM08.
The observational data (also in F? flux units) were
scaled to their apparent UKIDSS J-band magnitudes.
Simultaneous fits were made to the 0.85?1.35 ?m, 1.42?
1.8 ?m and 1.95?2.35 ?m regions to avoid strong tel-
luric absorption. For each data-model comparison, the
goodness-of-fitstatistic Gk (Cushing et al. 2008)was cal-
culated following the same weighting scheme employed
in that study, with each pixel weighted by its breadth
in wavelength space (wj ? ??j). Model surface fluxes
were scaled by the factor Ck ? (R/d)2 which minimizes
Gk (Equation 2 in Cushing et al. 2008), where R is the
radius of the brown dwarf and d its distance from the
Earth. As in Burgasser et al. (2008b), we generated dis-
tributionsofthe fitparametersbyweightingeachmodel?s
contribution according to the F-distribution probability
distribution function (PDF):
Wi ? 1?F(Gki/min(Gk) | ?eff,?eff). (1)
Here, Gki/min(Gk) is the ratio of goodness-of-fit (effec-
tively a ratio of chi-square residuals) between the best-fit
model and the ith model, and F(x|?eff,?eff) is the F-
distribution PDF for effective degrees of freedom
?eff ?
?
? 1
max({w})
Nsummationdisplay
j=1
wj
?
??1. (2)
The sum is over all pixels included in the fit (see
Burgasser et al. 2010a). The F-distribution provides a
Al2O3 (alumina or corundum). Other condensates are expected
to form above the silicate cloud at low effective temperature, but
of these only Na2S is likely to have any significant column opac-
ity (Marley 2000). In a future work we will consider the possible
contribution of this species.
Clouds in Ross 458C 9
more robust assessment of the equivalence of different
model fits than the exponential weighting employed in
Burgasser et al. (2008b) and Burgasser et al. (2010b).
Parametermeansandone-sidedstandarddeviationswere
separately computed as
?k? ?
summationtext
i Wikisummationtext
i Wi
(3)
and
?2k? =
summationtext
i? Wi(ki ??k?)
2
summationtext
i Wi
. (4)
The first sum was performed over all models and the
secondrestrictedto thosemodels whoseparametervalues
are above (+) or below (?) the weighted mean ?k?.
We examined several variations in our fitting method
to assess internal systematic effects. Modest changes to
the spectral regions that contributed to computing Gk
made little difference to the overall best fit. For ex-
ample, including the telluric regions in the fits, or ex-
panding the lower and upper ends of the spectral range
used, resulted in changes of less than 10 K and 0.1 dex
in the mean Teff or logg values for all three sources.
Only by excluding whole spectral regions?such as the
individual YJHK-band flux peaks?did inferred Teff
and logg values change significantly, by up to 100 K and
0.5 dex (similar to band-to-band variations reported in
Cushing et al. 2008). Variations in the pixel weighting
scheme also had negligible effect. Weighting all pixels
equally gave identical parameters (to within 10 K and
0.1dex in Teff and logg) asweighting by pixel bandpass.
Finally, using the exponential parameter weighting em-
ployed in Burgasser et al. (2010b), Wi ? e?0.5Gki, also
produced negligible changes in the inferred parameters
and their uncertainties. We therefore assume systematic
uncertainties of 20 K in Teff and 0.2 dex in logg, al-
though we cannot rule out larger systematic biases based
on the accuracy of the models.
4.2. Results
Table 3 summarizes the inferred parameters for each
source, with results for the cloudless and cloudy models
reported separately. Figures 5?7 show the best-fit mod-
els to the individual spectra, while Figures 8?10 show
the parameter distributions. It is first worth noting the
quality of the fits. For both cloudless and cloudy models,
deviations are seen in regions of strong CH4 absorption,
particularly in the 1.6-1.7 ?m region for Ross 458C and
ULAS J1335+1130 and the wings of the 1.27 ?m J-band
peak for SDSS J1416+1348B. As noted previously, these
discrepancies are likely the result of incomplete CH4
opacities (Saumon et al. 2006). However, there are also
differences between the cloudless and cloudy model fits,
notably in the relative YJH peak fluxes of Ross 458C
and ULAS J1335+1130. In the cloudless model fits of
these sources, the H-band peak appears to be suppressed
relative to J, while the Y-band peak appears suppressed
in the cloudy model fits. These differences are less pro-
nounced in fits to SDSS J1416+1348B. The cloudless
models also exhibit excessively strong K I lines at 1.24?
1.25 ?m, while these lines in the cloudy models are more
consistent with the data. For both models, the 2.1 ?m
K-band peak is reasonably well-reproduced for all three
sources (see also Cushing et al. 2008). Nevertheless, it
appears that the cloudy models provide better matches
to the data than the cloudless ones.
Quantitative differences between the cloudless and
cloudy model fits are also seen. For Ross 458C (Fig-
ures 5 and 8), the cloudy model provides a statisti-
cally superior fit to the data, significant at the 99.99%
confidence level based on the F-test PDF. The cloudy
model also predicts a significantly cooler temperature
than the cloudless case, Teff = 635+25?30 K versus Teff
= 760+70?45 K. The former value is consistent with the
luminosity-based Teff determined in Section 3.5, while
the latter is consistent with a ?normal? T8 field dwarf
like 2MASS J0415?0935 (e.g., Golimowski et al. 2004;
Stephens et al. 2009). Inferred surface gravities are com-
parablefor the two models, favoringlogg ?4.0 cgs, while
the cloudless models indicate a metal-rich atmosphere
([M/H] ? +0.3). These parameters are consistent with
the youth and metallicity of the Ross 458 system, as the
mean cloudless and cloudy model ages are 0.3+0.5?0.2 Gyr
and 0.14+0.03?0.01 Gyr, respectively. Both models alsopredict
a very low mass for Ross 458C, M = 0.010+0.006?0.004 M? and
M lessorsimilar0.006 M?, again at or below the deuterium burning
masslimit. Thereis a modestdifference between whether
vertical mixing is slightly favored (cloudless models) or
not (cloudy models), but it is not statistically signifi-
cant. The cloudless and cloudy models predict distinct
distances of 15.6?2.5 pc and 9.4?0.7 pc, respectively,
based on the optimal model scaling. These differ by 2.6?,
and are 1.8? higher and 2.4? lower than the parallactic
distance of the Ross 458 system, although the latter is
off by only 11%.
For ULAS J1335+1130 (Figures 6 and 9), the best-fit
cloudymodel alsohasa smallerGk value than the best-fit
cloudless one, although the statistical significance is re-
duced (84% confidence level). Again, the cloudy models
predict a significantly lower effective temperature than
the cloudless models (Teff = 595+25?45 K versus Teff =
700+30?50 K). The former is consistent with previously re-
ported values by Burningham et al. (2008, 550?600 K)
and Leggett et al. (2009, 500?550 K). The cloudy mod-
els also indicate a lower surface gravity (logg ? 4.0 cgs
versus logg = 4.7?0.4 cgs), while differences in vertical
mixing are statistically unconstrained. Like Ross 458C,
the cloudless models converge on a supersolar metallicity
for ULAS J1335+1130, [M/H] = +0.23+0.07?0.23. The differ-
ing parameters between the cloudless and cloudy models
translate into marginally overlapping ages (2.0+2.3?1.6 Gyr
versus 0.18+0.23?0.01 Gyr) and masses (0.010?0.032 M? ver-
sus 0.005?0.013M?), and distinct distances (20.5?3.6 pc
versus 14.4?1.1 pc, a 1.6? discrepancy).
Finally, for SDSS J1416+1348B (Figures 7 and 10),
we again find that the best-fit cloudy model is marginally
better than the best-fit cloudless one, albeit at even lower
statistical significance (confidence level of 77%) reflecting
the similarity in the near-infrared SEDs of these models.
Nevertheless, the inferred temperatures are marginally
distinct, with the cloudy models again indicating a cooler
temperature (Teff = 595+25?45 K versus Teff = 685+55?65 K).
Unlike Ross 458C and ULAS J1335+1130, the inferred
surface gravity of this source is considerably higher (logg
10 Burgasser et al.
1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4
Wavelength (?m)
0.0
0.2
0.4
0.6
0.8
f ? (mJy)
Ross 458C
700/4.0/+0.3/NC/104
d/R = 11.5 pc/RJup
Gk = 71.8
1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4
Wavelength (?m)
0.0
0.2
0.4
0.6
0.8
f ? (mJy)
Ross 458C
650/4.0/0.0/f2/0
d/R = 7.90 pc/RJup
Gk = 51.3
Figure 5. Best-fit cloudless (top) and cloudy (bottom) model fits (red lines) to the observed spectrum for Ross 458C (black lines).
Observed data are scaled to the apparent UKIDSS J-band magnitude of the source, while the model is scaled to minimize Gk (indicated,
along with d/R flux scaling factor). The noise spectrum for Ross 458C is shown by the grey lines, and fit regions indicated by the shaded
boxes at the top of each panel.
Clouds in Ross 458C 11
1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4
Wavelength (?m)
0.00
0.05
0.10
0.15
0.20
0.25
0.30
f ? (mJy)
ULAS J1335+1130
700/5.0/+0.3/NC/104
d/R = 20.9 pc/RJup
Gk = 22.7
1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4
Wavelength (?m)
0.00
0.05
0.10
0.15
0.20
0.25
0.30
f ? (mJy)
ULAS J1335+1130
600/4.0/0.0/f2/0
d/R = 12.0 pc/RJup
Gk = 21.0
Figure 6. Same as Figure 5 for ULAS J1335+1130.
12 Burgasser et al.
1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4
Wavelength (?m)
0.0
0.1
0.2
0.3
0.4
f ? (mJy)
SDSS J1416+1348B
700/5.5/0.0/NC/104
d/R = 15.0 pc/RJup
Gk = 22.3
1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4
Wavelength (?m)
0.0
0.1
0.2
0.3
0.4
f ? (mJy)
SDSS J1416+1348B
600/5.5/0.0/f2/0
d/R = 15.1 pc/RJup
Gk = 21.0
Figure 7. Same as Figure 5 for SDSS J1416+1348B.
Clouds in Ross 458C 13
Teff (K)
0.0
0.2
0.4
0.6
0.8
1.0
1.2
Normalized Distribution
500 600 700 800 900 1000
log g (cgs)
0.0
0.2
0.4
0.6
0.8
1.0
1.2
Normalized Distribution
4.0 4.5 5.0 5.5
Kzz (cm2 s-1)
0.0
0.2
0.4
0.6
0.8
1.0
1.2
Normalized Distribution
0 104
[M/H] (dex)
0.0
0.2
0.4
0.6
0.8
1.0
1.2
Normalized Distribution
-0.3 0 +0.3
Teff (K)
0.0
0.2
0.4
0.6
0.8
1.0
1.2
Normalized Distribution
500 600 700 800 900 1000
log g (cgs)
0.0
0.2
0.4
0.6
0.8
1.0
1.2
Normalized Distribution
4.0 4.5 5.0 5.5
Kzz (cm2 s-1)
0.0
0.2
0.4
0.6
0.8
1.0
1.2
Normalized Distribution
0 104
Figure 8. Parameter distributions for cloudless (top) and cloudy (bottom) model fits of Ross 458C, based on the parameter weighting
scheme described in the text. From left to right, distributions in Teff, logg and Kzz are shown. For the cloudless models, the distribution
in [M/H] values is also shown.
Teff (K)
0.0
0.2
0.4
0.6
0.8
1.0
1.2
Normalized Distribution
500 600 700 800 900 1000
log g (cgs)
0.0
0.2
0.4
0.6
0.8
1.0
1.2
Normalized Distribution
4.0 4.5 5.0 5.5
Kzz (cm2 s-1)
0.0
0.2
0.4
0.6
0.8
1.0
1.2
Normalized Distribution
0 104
[M/H] (dex)
0.0
0.2
0.4
0.6
0.8
1.0
1.2
Normalized Distribution
-0.3 0 +0.3
Teff (K)
0.0
0.2
0.4
0.6
0.8
1.0
1.2
Normalized Distribution
500 600 700 800 900 1000
log g (cgs)
0.0
0.2
0.4
0.6
0.8
1.0
1.2
Normalized Distribution
4.0 4.5 5.0 5.5
Kzz (cm2 s-1)
0.0
0.2
0.4
0.6
0.8
1.0
1.2
Normalized Distribution
0 104
Figure 9. Same as Figure 8 for ULAS J1335+1130.
14 Burgasser et al.
Teff (K)
0.0
0.2
0.4
0.6
0.8
1.0
1.2
Normalized Distribution
500 600 700 800 900 1000
log g (cgs)
0.0
0.2
0.4
0.6
0.8
1.0
1.2
Normalized Distribution
4.0 4.5 5.0 5.5
Kzz (cm2 s-1)
0.0
0.2
0.4
0.6
0.8
1.0
1.2
Normalized Distribution
0 104
[M/H] (dex)
0.0
0.2
0.4
0.6
0.8
1.0
1.2
Normalized Distribution
-0.3 0 +0.3
Teff (K)
0.0
0.2
0.4
0.6
0.8
1.0
1.2
Normalized Distribution
500 600 700 800 900 1000
log g (cgs)
0.0
0.2
0.4
0.6
0.8
1.0
1.2
Normalized Distribution
4.0 4.5 5.0 5.5
Kzz (cm2 s-1)
0.0
0.2
0.4
0.6
0.8
1.0
1.2
Normalized Distribution
0 104
Figure 10. Same as Figure 8 for SDSS J1416+1348B.
= 5.2?0.3cgs cloudless; logg ?5.5 cgscloudy), while the
cloudless model indicates a subsolar metallicity ([M/H]
= -0.17+0.17?0.13). As discussed in Burningham et al. (2010)
and Burgasser et al. (2010b), high surface gravityand/or
subsolarmetallicityarerequiredtoproduceboth the sup-
pressed K-band peak and broadened Y-band peak ob-
served in the spectrum of this source. Vertical mixing
is again poorly constrained. The inferred atmospheric
parameters of SDSS J1416+1348B indicate a brown
dwarf that is older (4?12 Gyr) and more massive (0.03?
0.05 M?) than the other two sources, consistent with
prior analyses. Surprisingly, the cloudless and cloudy
models convergeona commondistanceof11?3pc, larger
than but formally consistent with the preliminary paral-
lax measurement of Scholz (2010b, 7.9?1.7 pc).
5. DISCUSSION
5.1. Clouds in Young, Cold and Metal-Rich T Dwarfs
The above analysis illustrates how condensate opac-
ity, when included in spectral models, can have a pro-
found effect on the near-infrared spectral energy dis-
tributions and inferred Teffs of the coldest T dwarfs.
More importantly, the inclusion of condensate opacity
provides a statistically significant improvement in fits to
the near-infrared spectrum of Ross 458C, and marginally
so for ULAS J1335+1130 and SDSS J1416+1348B. This
is the notthe firstindication ofcondensatesbeing present
in T dwarf atmospheres. Early-type T dwarfs exhibit
evidence of waning cloud opacity (Marley et al. 2002;
Cushing et al. 2008; Leggett et al. 2008), and SM08 have
shown that T dwarf models produce excessively blue
near-infrared colors without the inclusion of condensate
extinction. Nevertheless, several theoretical studies have
concluded that condensate clouds play little or no role in
shaping the near-infrared spectra of brown dwarfs cooler
than ?1000 K, as they reside below the visible photo-
sphere (e.g., Allard et al. 2001; Burrows et al. 2006). In-
stead, we find that clouds are essential for reproducing
the near-infrared SED of the Teff ? 650 K T dwarf
Ross 458C, and to a lesser degree the Teff ? 600 K T
dwarfs ULAS J1335+1130 and SDSS J1416+1348B.
The importance of clouds in these two sources is
likely related to their youth and associated low sur-
face gravities. Among the L dwarfs, a correlation
between low surface gravity and thick clouds has
been noted by Cushing et al. (2008) and Stephens et al.
(2009) based on broad-band spectral model fits; and by
Metchev & Hillenbrand (2006) and Dupuy et al. (2009b)
based on the delayed disruption of clouds at the L
dwarf/T dwarf transition. There is also evidence that
this correlation continues into the T dwarf regime, based
on cloudy model fits to broad-band spectral data for
the low-gravity T5.5 dwarf SDSSp J111010.01+011613.1
(Stephens et al. 2009). Finally, the 9?11 ?m silicate
grain feature has also been shown to be more pro-
nounced in L dwarfs with unusually red J ? K col-
ors (Burgasser et al. 2008a; Looper et al. 2008), sources
whose kinematics reflect youth and low surface gravities
(Faherty et al. 2009; Schmidt et al. 2010).
Atmospheric models show that two complementary ef-
fects can drive the persistence of clouds in low gravity,
low-temperature brown dwarf atmospheres. First, the
column of atmosphere above a given pressure level is
deeper in a lower gravity dwarf, and as a consequence
the atmosphere is everywhere warmer at a fixed Teff.
This implies that silicate clouds lie higher in these at-
mospheres, providing a larger column depth of extinc-
tion in regions of minimum molecular gas opacity. Sec-
ond, cloud models from Ackerman & Marley (2001) and
Helling et al. (2008) both predict that mean cloud par-
ticle sizes are larger in lower gravity atmospheres. In a
Teff = 650 K SM08 model, for example, the mean par-
ticle radius is roughly ten times larger for logg = 4.5
as compared to logg = 5.5. The sub-micron particles
found in higher gravity models contribute little opac-
Clouds in Ross 458C 15
Table 3
Model-fit Parameters for Observed T Dwarfs.
Parameter Best Fit Mean Best Fit Mean Prior Value Ref.
(Cloudless) (Cloudless) (Cloudy) (Cloudy)
Ross 458C (T8)
Gk 71.8 ??? 51.3 ??? ???
d/R (pc/RJup) 11.5 13.4?1.9 7.9 7.5?0.6 ???
Teff (K) 700 760+70?45 650 635+25?35 ???
logg (cm s?2) 4.0 4.2+0.3?0.2 4.0 ?4.0 ???
[M/H] (dex) +0.3 ?+0.3 0a 0a +0.2...+0.3 1
Kzz (cm2 s?1) 104 ??? 0 ??? ???
Mass (M?) 0.006 0.010+0.006?0.004 0.006 lessorsimilar0.006 0.006-0.011b 1
Age (Gyr) 0.08 0.3+0.5?0.2 0.13 0.14+0.03?0.01 0.4?0.8 1
0.4-2 2
R (RJup) 1.22 1.16+0.07?0.09 1.25 ?1.24 ???
d (pc) 14.1 15.6?2.5 9.8 9.4?0.7 11.7?0.2 3
ULAS J1335+1130 (T9)
Gk 22.7 ??? 21.0 ??? ???
d/R (pc/RJup) 20.9 20.2?2.4 12.0 11.8?0.8 ???
Teff (K) 700 700+30?50 600 595+25?45 500-600 4,5
logg (cm s?2) 5.0 4.7?0.4 4.0 ?4.0 4.0?4.5 4,5
[M/H] (dex) +0.3 +0.23+0.07?0.23 0a 0a 0...+0.3 5
Kzz (cm2 s?1) 104 ??? 0 ??? ???
Mass (M?) 0.031 0.021?0.011 0.006 0.005?0.013 0.014?0.030 4
0.005-0.019 5
Age (Gyr) 3.3 2.0+2.3?1.6 0.17 0.18+0.23?0.01 0.6?5.3 4
0.01-2 5
R (RJup) 0.90 1.01+0.13?0.12 1.23 ?1.23 1.07?1.17 5
d (pc) 18.7 20.5?3.6 14.7 14.4?1.1 8?12 4,5
SDSS J1416+1348B (T7.5p)
Gk 22.3 ??? 21.0 ??? ???
d/R (pc/RJup) 15.0 13.2?3.6 15.1 13.2?4.0 12.8?3.0 6
Teff (K) 700 685+55?65 600 595+25?45 650?60 6
500 7
logg (cm s?2) 5.5 5.2?0.3 5.5 ?5.5 5.2?0.4 6,7
[M/H] (dex) 0.0 -0.17+0.17?0.13 0a 0a -0.3 6,7
Kzz (cm2 s?1) 104 ??? 0 ??? 104 6,7
Mass (M?) 0.052 0.041+0.011?0.010 0.036 0.035+0.001?0.003 0.021?0.045 6
0.029?0.038 7
Age (Gyr) 12 8?4 12 6?12 2?10 6,7
R (RJup) 0.78 0.84?0.06 0.86 ?0.86 0.83+0.14?0.10 6
d (pc) 11.8 11.1?3.2 13.0 11.4?3.4 7.9?1.7 8
References. ? (1) This paper; (2) Goldman et al. (2010); (3) van Leeuwen (2007);
(4) Burningham et al. (2008); (5) Leggett et al. (2009); (6) Burgasser et al. (2010b);
(7) Burningham et al. (2010); (8) Scholz (2010b).
a Solar metallicity is assumed for all cloudy models.
b Mass estimate based on the luminosity and estimated age of Ross 458C, and the
cloudy evolutionary models of SM08.
ity in the near-infrared because of Mie-scattering effects,
rendering the deep silicate cloud less visible. Clouds in
lower gravity dwarfs are therefore more visible and scat-
ter more efficiently than those in their older and more
massive counterparts. This interpretation explains why
cloudless models have so far provided adequate fits to the
spectra of logg greaterorsimilar 5 T dwarfs (e.g., Geballe et al. 2001;
Saumon et al. 2007; Burgasser et al. 2008b), including
SDSS J1416+1348B.
Metallicity is another factor to consider. Cloud for-
mation appears to be inhibited in metal-poor L sub-
dwarfs, as exemplified by the retention of Ca, Ti and
TiO gas to cooler Teffs in these objects (Burgasser et al.
2003, 2007; Gizis & Harvin 2006). Similarly, unusu-
ally blue L dwarfs ? including the brighter compan-
ion of SDSS J1416+1348B ? appear to have slightly
metal-poor atmospheres with unusually thin clouds
(Burgasser et al. 2008a; Cushing et al. 2010) or frac-
tured cloud layers (Marley et al. submitted). It is not
unreasonable to speculate that supersolar metallicity
dwarfs may have thicker clouds (Looper et al. 2008), yet
theoretical predictions are mixed. SM08 have shown
that warmer cloudy models (Teff ? 1300 K) appear
?cloudier? (redder) for supersolar metallicities. How-
ever, chemical equilibrium calculations by Visscher et al.
(2010) find that MgSiO3 (enstatite) condenses at warmer
temperatures in higher metalllicity atmospheres, leading
to a deeper cloud that may be buried by gas opacity
in cooler brown dwarfs. Atmospheric model calculations
by Witte et al. (2009) find that grain size and cloud op-
tical depth do not change appreciably between [M/H] =
0 and +0.5 for low-temperature brown dwarfs, although
the clouds themselves expand into lower pressure regions
of the atmosphere and may have a greater effect at wave-
16 Burgasser et al.
lengths corresponding to minimum gas opacity. Metal-
licity effects on cloud formation and cloud properties in
cool T dwarf atmospheres clearly remains an area of fur-
ther theoretical exploration.
In summary, the unexpected presence of condensate
absorption in the spectrum of Ross 458C (and possi-
bly ULAS J1335+1130) can be attributed to its low
surface gravity, and possibly its supersolar metallic-
ity. This hypothesis follows similar trends observed in
the L dwarfs. It also explains why clouds had not
previously been needed in model fits of high surface
gravity T dwarfs. The importance of clouds in shap-
ing low gravity T dwarf spectra should nevertheless
be a consideration in the characterization of planetary-
mass brown dwarf candidates in local young clusters
(e.g., Zapatero Osorio et al. 2002; Casewell et al. 2007;
Burgess et al. 2009; Marsh et al. 2010) and low-mass
exoplanets recently imaged around nearby young stars
(e.g., Kalas et al. 2008; Marois et al. 2008; Janson et al.
2010; Lagrange et al. 2010).
5.2. Clouds and the Convergence of Near-infrared and
Mid-infrared T Dwarf Temperature Scales
Spitzer has enabled detailed studies of cold T dwarfs in
the mid-infrared, where the majority of the SED emerges
(Leggett et al. 2010). Model fits to mid-infrared spectral
data have proven particularly robust, with agreement
in continuum fluxes and prominent H2O, CH4 and NH3
bands (e.g., Saumon et al. 2007; Burgasser et al. 2008b;
Leggett et al. 2009). However, there have been indica-
tions that temperatures inferred from near- and mid-
infrared analyses can differ significantly. A case in point
is SDSS J1416+1348B. Burgasser et al. (2010b) inferred
Teff = 650?60 K for this source from near-infrared
cloudless spectral model fits, similar to the cloudless re-
sults presented here. Burningham et al. (2010) inferred
Teff ? 500 K based on the extreme H ? [4.5] color of
this source (Warren et al. 2007; Leggett et al. 2010). We
find that inclusion of cloud opacity in the near-infrared
modeling of this source brings inferred Teffs closer to
mid-infrared results, despite having minimal effect on the
shape of its spectrum. Even better convergence is seen
forULAS J1335+1130,asournear-infraredcloudymodel
temperature is fully consistent with estimates based on
H ? [4.5] color (Burningham et al. 2008) and near- to
mid-infrared spectral model fits (Leggett et al. 2009).
For Ross 458C, the cloudy model temperature is consis-
tent with that inferred from evolutionary models based
on the age and luminosity of the source (Section 3.5).
Prior mid-infrared analyses have been based on cloud-
less model fits, however, so does the inclusion of clouds
affect these results? To explore this question, Fig-
ure 11 compares the observed mid-infrared spectrum of
ULAS J1335+1130 obtained with the Spitzer Infrared
Spectrograph (IRS; Houck et al. 2004; Leggett et al.
2009) to three SM08 models: two sampling the best-fit
parameters from our near-infrared cloudless and cloudy
model fits, and one sampling the cloudless model fit
parameters from Leggett et al. (2009).18 The first two
models are scaled according to the near-infrared nor-
18 Leggett et al. (2009) used Kzz = 106 cm2 s?1 in their best-
fit models whereas we use Kzz = 104 cm2 s?1 in Figure 11. The
difference between these models is negligible.
6 8 10 12 14
Wavelength (?m)
0.0
0.2
0.4
0.6
0.8
f ? (mJy)
550/4.0/0.0/104
700/5.0/+0.3/104
600/4.0/f2/0 (104)
ULAS J1335+1130
Figure 11. Spitzer/IRS spectrum of ULAS J1335+1130 from
Leggett et al. (2009, black line, noise spectrum in grey), scaled
to its apparent 8.0 ?m magnitude (Burningham et al. 2008), com-
pared to best-fit cloudless (blue line) and cloudy (solid red line)
models from our near-infrared analysis, and the best-fit cloudless
model from the near- and mid-infrared analysis of Leggett et al.
(2009, green line). Model surface fluxes have been smoothed to
the resolution of the IRS data (?/?? ? 100) using a Gaussian
kernel, and scaled according to their respective best-fit distances
and radii. The best-fit cloudy model to the near-infrared data with
Kzz = 104 cm2 s?1 is also shown (dashed red line). Observational
uncertainty in flux at 10 ?m in indicated by the error bar.
malizations from our analysis, while the Leggett et al.
model is scaled to the best-fit radius and distance de-
rived in that study. The 600 K cloudy model comes much
closer to reproducing the pseudo-continuum flux of the
observed data than the 700 K cloudless model, but is still
underluminous and predicts excessively strong NH3 ab-
sorption in the 9?13?m region. This is not a cloud effect,
however, but the absence of vertical mixing in the best-
fit cloudy model. Vertical mixing reduces NH3 and CO
photosphericabundances through non-equilibrium chem-
istry, but has little impact on the near-infrared spectrum
for Teff lessorsimilar 1000 K (Griffith & Yelle 1999; Saumon et al.
2006). This is confirmed in our spectral modeling, as the
presence or absence of vertical mixing is statistically am-
biguous (Figures 8?10). It has a profound effect on mid-
infrared spectra, however, as the Teff = 600 K, logg =
4.0, Kzz = 104 cm2 s?1 cloudy model is a good match to
data for ULAS J1335+1130 (i.e., consistent within the
observational uncertainties). As such, the inclusion of
cloud opacity does not appear to modify mid-infrared
analyses for the coldest T dwarfs.
Convergence between near-infrared and mid-infrared
spectral model fit parameters and temperature scales for
cold T dwarfs is important for both near- and long-
term study of these objects. With Spitzer now de-
pleted of cryogen, our ability to obtain mid-infrared
spectroscopy for faint brown dwarfs is restricted pend-
ing deployment of the James Webb Space Telescope
(JWST) Mid-Infrared Instrument (MIRI; Wright et al.
2008) later this decade. In addition, characterization
of several dozens of cold brown dwarf discoveries ex-
pected from UKIDSS, the Canada France Hawaii Tele-
scope Legacy Survey (CFHTLS; Delorme et al. 2008),
Clouds in Ross 458C 17
the Wide Field InfraredExplorer(WISE; Liu et al. 2008)
and the Visible and Infrared Survey Telescope for As-
tronomy (VISTA; McPherson et al. 2006) cannot rely on
space-basedspectroscopy. Rather, accuratereproduction
of ground-based near-infrared spectroscopy, facilitated
by sensitive instruments such as FIRE, will be necessary
to study the physical characteristics and demographics
of these intrinsically faint sources.
5.3. Is Ross 458C a Planet?
Finally, we consider whether the properties of
Ross 458C qualify it as a ?planet? (Basri & Brown 2006;
Soter 2006; Marchi 2007). In terms of characteristics,
the planet moniker appears appropriate. Ross 458C is
gravitationally bound to stellar (hydrogen-fusing) com-
panions, and itself appears to have a mass below which
no fusion is expected. Its relative mass is roughly 1:50
compared to its companions. Its surface temperature is
well below those of several well-studied, so-called plan-
ets; e.g., AB Pic b (2000+100?300 K; Bonnefoy et al. 2010),
1RXS J160929.1-210524b (1800?200K;Lafreni`ere et al.
2010), 2MASS 1207-3932b (1600?100 K; Mohanty et al.
2007), and HR 8799 b (1300?1700K, Bowler et al. 2010).
Evidence of clouds also lends Ross 458C a decidedly
planetary-like character.
However, cosmogony seems to indicate otherwise.
Ross 458C?s 1100 AU projected separation makes it
highly unlikely to have formed according to the ?stan-
dard? model; i.e., the accumulation of gas and dust
in a circumstellar (perhaps in this case circumbinary)
disk (Pollack et al. 1996). Indeed, if Ross 458C is a
planet, it has the widest and longest (?50,000 yr) plan-
etary orbit known, 2-4 times wider than AB Pic b
(M ? 0.011 M?, a ? 260 AU; Chauvin et al. 2005;
Bonnefoy et al. 2010), 1RXS J160929.1-210524 (M ?
0.008 M?, a ? 330 AU; Lafreni`ere et al. 2008, 2010), CT
Cha b (M = 0.016?0.006AU, a?440AU; Schmidt et al.
2008) and UScoCTIO 106b (M = 0.013+0.002?0.008, a ?
670 AU; B?ejar et al. 2008). Can planets form at these
separations? Some disk fragmentation simulations have
been able to produce very wide (>1000 AU) substel-
lar companions, albeit with extreme initial conditions
(e.g., Mdisk/M? ? 1; Stamatellos & Whitworth 2009;
Vorobyov & Basu 2010; see also Boss 2006). Planet-
planet interactions have been shown to scatter sources
out to orbits 100?1000 AU in size, although they are ex-
ceedingly rare (<1%) and short-lived in simulations (e.g.,
Dodson-Robinson et al. 2009; Veras et al. 2009). The
triple nature of Ross 458 may provide an angular mo-
mentum sink for such a scenario, allowing a planet to be
scattered out without being ejected (Holman & Wiegert
1999; Nelson 2003; Pierens & Nelson 2008). All of these
scenarios require dramatic?and highly fortuitous?
dynamic interactions to send a planet formed in the re-
gion of a circumstellar/circumbinary disk to an orbital
radius 100 times wider than Jupiter?s. Of course, a sim-
pler alternative may be that Ross 458 formed as a ?nor-
mal? hierarchical triple star system, albeit one with very
low mass components.
Ross 458C therefore stands as a benchmark not just for
studies of low-temperature, planetary-like atmospheres,
but for the definition of the word planet in general. Bet-
ter characterization of the age of this system and the
orbital and physical properties of its components will ul-
timately help us address what parameter space defines a
planet, and whether ?well-defined boundaries, empirical
verifiability, or logical consistency ...[can] overcome gut
feelings? (Basri & Brown 2006).
6. SUMMARY
We have obtained FIRE near-infrared spectra of the
faint companion Ross 458C and the late-type field T
dwarfs ULAS J1335+1130 and SDSS J1416+1348B.
Data for the first source confirm its substellar nature and
identify it as a T8 dwarf, with an inferred log10 Lbol/L?
= ?5.62?0.03 and Teff = 650?25 K based on empirical
bolometric corrections and evolutionary models. These
data also show a pronounced K-band peak and relatively
red J?K spectrophotometric colors, indicative of super-
solar metallicity and youth that coincide with the prop-
erties of the Ross 458system (age = 150?800Myr, [Fe/H]
= +0.2 to +0.3). Fits of the data to synthetic spectra
from SM08 confirm these secondary parameters, albeit
with discrepant effective temperatures and distances de-
pending on the presence or absence of condensate opac-
ity. In fact, cloudy models provide significantly better
fits to the near-infraredspectrum ofRoss 458Cand closer
agreementwith its parallaxdistance, while alsoproviding
marginally better fits to data for ULAS J1335+1130 and
SDSS J1416+1348B. The fact that cloud opacity effects
are more pronounced in the spectra of Ross 458C and
ULAS J1335+1130 indicates that clouds are particularly
relevant for shaping the spectra of young, low-gravity
brown dwarfs, and/or those with supersolar metallicities,
following trends seen in the warmer L dwarfs.
Despite the improvement, model fits to brown dwarf
spectra remain a work in progress. The cloudy mod-
els ?overcorrect? Y- and J-band fluxes and underesti-
mate absolute fluxes in the case of Ross 458C. Both
shortcomings suggest that thinner clouds (e.g., fsed =
3 or 4) may be needed to accurately account for con-
densate extinction, in line with values inferred for L/T
transition objects (Cushing et al. 2008; Stephens et al.
2009). Furthermore, the limited sample examined here
may not be representative of the greater population of
low-temperature, low surface gravity T dwarfs. A com-
prehensive modeling effort can more fully explore the
properties of clouds in these objects, while accurate par-
allax measurements would provide the necessary empir-
ical constraints on surface fluxes. Nevertheless, our re-
sults indicate that by including condensate opacity in
T dwarf spectral models, more accurate determinations
of the physical parameters of cold brown dwarfs and
exoplanets are possible with near-infrared spectroscopy
alone.
The authors acknowledges the efforts of the Las Cam-
panas Observatory staff in helping us bring FIRE to the
first light and science operations. Specifically, we thank
Frank Per?ez, Alan Uomoto, Mark Phillips, Dave Osip,
and Povilas Palunas for their assistance during commis-
sioning. We also thank our telescope operators Jorge
Araya, Mauricio Martinez, and Hern?an Nu?nez for their
assistance during the observations. We acknowledge
helpful comments from G. Bjorn, B. Bowler, J. Faherty,
J. Gizis, D. Looper and R.-D. Scholz; and thank our ref-
eree Channon Visscher for his prompt and helpful review
18 Burgasser et al.
of our original manuscript. The authors also thank B.
Burningham and S. Leggett for providing their spectral
dataofULAS J1335+1130,and J. D. Kirkpatrickforpro-
viding his spectrum of Ross 458AB used in the analysis.
Support for the modeling work of D.S. was provided by
NASA through the Spitzer Science Center. This research
has benefitted from the M, L, and T dwarf compendium
housed at DwarfArchives.org and maintained by Chris
Gelino, Davy Kirkpatrick, and Adam Burgasser.
Facilities: Magellan (FIRE)
REFERENCES
Ackerman, A. S., & Marley, M. S. 2001, ApJ, 556, 872
Allard, F., Hauschildt, P. H., Alexander, D. R., Tamanai, A., &
Schweitzer, A. 2001, ApJ, 556, 357
Allard, N. F., Allard, F., Hauschildt, P. H., Kielkopf, J. F., &
Machin, L. 2003, A&A, 411, L473
Antoja, T., Figueras, F., Fern?andez, D., & Torra, J. 2008, A&A,
490, 135
Barnes, S. A. 2003, ApJ, 586, 464
Basri, G., & Brown, M. E. 2006, Annual Review of Earth and
Planetary Sciences, 34, 193
B?ejar, V. J. S., Zapatero Osorio, M. R., P?erez-Garrido, A.,
?Alvarez, C., Mart??n, E. L., Rebolo, R., Vill?o-P?erez, I., &
D??az-S?anchez, A. 2008, ApJ, 673, L185
Beuzit, J.-L., S?egransan, D., Forveille, T., Udry, S., Delfosse, X.,
Mayor, M., Perrier, C., Hainaut, M.-C., Roddier, C., Roddier,
F., & Mart??n, E. L. 2004, A&A, 425, 997
Bildsten, L., Brown, E. F., Matzner, C. D., & Ushomirsky, G.
1997, ApJ, 482, 442
Bochanski, J. J., West, A. A., Hawley, S. L., & Covey, K. R.
2007, AJ, 133, 531
Bonfils, X., Delfosse, X., Udry, S., Santos, N. C., Forveille, T., &
S?egransan, D. 2005, A&A, 442, 635
Bonnefoy, M., Chauvin, G., Rojo, P., Allard, F., Lagrange, A.,
Homeier, D., Dumas, C., & Beuzit, J. 2010, A&A, 512, A52+
Borysow, A. 2002, A&A, 390, 779
Boss, A. P. 2006, ApJ, 637, L137
Bowler, B. P., Liu, M. C., & Cushing, M. C. 2009, ApJ, 706, 1114
Bowler, B. P., Liu, M. C., Dupuy, T. J., & Cushing, M. C. 2010,
ArXiv e-prints
Browning, M. K., Basri, G., Marcy, G. W., West, A. A., &
Zhang, J. 2010, AJ, 139, 504
Burgasser, A. J. 2007, ApJ, 659, 655
Burgasser, A. J., Burrows, A., & Kirkpatrick, J. D. 2006a, ApJ,
639, 1095
Burgasser, A. J., Cruz, K. L., Cushing, M., Gelino, C. R., Looper,
D. L., Faherty, J. K., Kirkpatrick, J. D., & Reid, I. N. 2010a,
ApJ, 710, 1142
Burgasser, A. J., Cruz, K. L., & Kirkpatrick, J. D. 2007, ApJ,
657, 494
Burgasser, A. J., Geballe, T. R., Leggett, S. K., Kirkpatrick,
J. D., & Golimowski, D. A. 2006b, ApJ, 637, 1067
Burgasser, A. J., Kirkpatrick, J. D., Brown, M. E., Reid, I. N.,
Burrows, A., Liebert, J., Matthews, K., Gizis, J. E., Dahn,
C. C., Monet, D. G., Cutri, R. M., & Skrutskie, M. F. 2002a,
ApJ, 564, 421
Burgasser, A. J., Kirkpatrick, J. D., Burrows, A., Liebert, J.,
Reid, I. N., Gizis, J. E., McGovern, M. R., Prato, L., &
McLean, I. S. 2003, ApJ, 592, 1186
Burgasser, A. J., Looper, D., & Rayner, J. T. 2010b, AJ, 139,
2448
Burgasser, A. J., Looper, D. L., Kirkpatrick, J. D., Cruz, K. L., &
Swift, B. J. 2008a, ApJ, 674, 451
Burgasser, A. J., Marley, M. S., Ackerman, A. S., Saumon, D.,
Lodders, K., Dahn, C. C., Harris, H. C., & Kirkpatrick, J. D.
2002b, ApJ, 571, L151
Burgasser, A. J., Tinney, C. G., Cushing, M. C., Saumon, D.,
Marley, M. S., Bennett, C. S., & Kirkpatrick, J. D. 2008b, ApJ,
689, L53
Burgasser, A. J., et al. 2000, ApJ, 531, L57
Burgess, A. S. M., Moraux, E., Bouvier, J., Marmo, C., Albert,
L., & Bouy, H. 2009, A&A, 508, 823
Burningham, B., Leggett, S. K., Lucas, P. W., Pinfield, D. J.,
Smart, R. L., Day-Jones, A. C., Jones, H. R. A., Murray, D.,
Nickson, E., Tamura, M., Zhang, Z., Lodieu, N., Tinney, C. G.,
& Zapatero Osorio, M. R. 2010, MNRAS, 404, 1952
Burningham, B., et al. 2008, MNRAS, 391, 320
?. 2009, MNRAS, 395, 1237
Burrows, A., Burgasser, A. J., Kirkpatrick, J. D., Liebert, J.,
Milsom, J. A., Sudarsky, D., & Hubeny, I. 2002, ApJ, 573, 394
Burrows, A., Sudarsky, D., & Hubeny, I. 2006, ApJ, 640, 1063
Burrows, A., & Volobuyev, M. 2003, ApJ, 583, 985
Casewell, S. L., Dobbie, P. D., Hodgkin, S. T., Moraux, E.,
Jameson, R. F., Hambly, N. C., Irwin, J., & Lodieu, N. 2007,
MNRAS, 378, 1131
Chabrier, G., & Baraffe, I. 2000, ARA&A, 38, 337
Chabrier, G., Baraffe, I., Allard, F., & Hauschildt, P. 2000, ApJ,
542, 464
Chabrier, G., Baraffe, I., & Plez, B. 1996, ApJ, 459, L91+
Chauvin, G., Lagrange, A.-M., Zuckerman, B., Dumas, C.,
Mouillet, D., Song, I., Beuzit, J.-L., Lowrance, P., & Bessell,
M. S. 2005, A&A, 438, L29
Cooper, C. S., Sudarsky, D., Milsom, J. A., Lunine, J. I., &
Burrows, A. 2003, ApJ, 586, 1320
Cushing, M. C., Marley, M. S., Saumon, D., Kelly, B. C., Vacca,
W. D., Rayner, J. T., Freedman, R. S., Lodders, K., & Roellig,
T. L. 2008, ApJ, 678, 1372
Cushing, M. C., Rayner, J. T., & Vacca, W. D. 2005, ApJ, 623,
1115
Cushing, M. C., Roellig, T. L., Marley, M. S., Saumon, D.,
Leggett, S. K., Kirkpatrick, J. D., Wilson, J. C., Sloan, G. C.,
Mainzer, A. K., Van Cleve, J. E., & Houck, J. R. 2006, ApJ,
648, 614
Cushing, M. C., Saumon, D., & Marley, M. S. 2010, ArXiv
e-prints
Cushing, M. C., Vacca, W. D., & Rayner, J. T. 2004, PASP, 116,
362
de Bruijne, J. H. J., Hoogerwerf, R., & de Zeeuw, P. T. 2001,
A&A, 367, 111
Dehnen, W., & Binney, J. J. 1998, MNRAS, 298, 387
Delorme, P., Willott, C. J., Forveille, T., Delfosse, X., Reyl?e, C.,
Bertin, E., Albert, L., Artigau, E., Robin, A. C., Allard, F.,
Doyon, R., & Hill, G. J. 2008, A&A, 484, 469
Dodson-Robinson, S. E., Veras, D., Ford, E. B., & Beichman,
C. A. 2009, ApJ, 707, 79
Dupuy, T. J., Liu, M. C., & Ireland, M. J. 2009a, ApJ, 692, 729
?. 2009b, ApJ, 699, 168
Eggen, O. J. 1960, MNRAS, 120, 540
Faherty, J. K., Burgasser, A. J., Cruz, K. L., Shara, M. M.,
Walter, F. M., & Gelino, C. R. 2009, AJ, 137, 1
Famaey, B., Jorissen, A., Luri, X., Mayor, M., Udry, S.,
Dejonghe, H., & Turon, C. 2005, A&A, 430, 165
Feast, M., & Whitelock, P. 1997, MNRAS, 291, 683
Fegley, B. J., & Lodders, K. 1996, ApJ, 472, L37+
Fleming, T. A., Schmitt, J. H. M. M., & Giampapa, M. S. 1995,
ApJ, 450, 401
Freedman, R. S., Marley, M. S., & Lodders, K. 2008, ApJS, 174,
504
Geballe, T. R., Knapp, G. R., Leggett, S. K., Fan, X.,
Golimowski, D. A., Anderson, S., Brinkmann, J., Csabai, I.,
Gunn, J. E., Hawley, S. L., Hennessy, G., Henry, T. J., Hill,
G. J., Hindsley, R. B., Ivezi?c, ?Z., Lupton, R. H., McDaniel, A.,
Munn, J. A., Narayanan, V. K., Peng, E., Pier, J. R., Rockosi,
C. M., Schneider, D. P., Smith, J. A., Strauss, M. A.,
Tsvetanov, Z. I., Uomoto, A., York, D. G., & Zheng, W. 2002,
ApJ, 564, 466
Geballe, T. R., Saumon, D., Leggett, S. K., Knapp, G. R.,
Marley, M. S., & Lodders, K. 2001, ApJ, 556, 373
Gizis, J. E., & Harvin, J. 2006, AJ, 132, 2372
Goldman, B., Marsat, S., Henning, T., Clemens, C., & Greiner, J.
2010, MNRAS, 405, 1140
Golimowski, D. A., et al. 2004, AJ, 127, 3516
Greiner, J., Bornemann, W., Clemens, C., Deuter, M., Hasinger,
G., Honsberg, M., Huber, H., Huber, S., Krauss, M., Kr?uhler,
T., K?upc?u Yolda?s, A., Mayer-Hasselwander, H., Mican, B.,
Primak, N., Schrey, F., Steiner, I., Szokoly, G., Th?one, C. C.,
Yolda?s, A., Klose, S., Laux, U., & Winkler, J. 2008, PASP, 120,
405
Griffith, C. A., & Yelle, R. V. 1999, ApJ, 519, L85
Clouds in Ross 458C 19
Helling, C., Dehn, M., Woitke, P., & Hauschildt, P. H. 2008, ApJ,
675, L105
Helling, C., Thi, W.-F., Woitke, P., & Fridlund, M. 2006, A&A,
451, L9
Hodapp, K. W., Jensen, J. B., Irwin, E. M., Yamada, H., Chung,
R., Fletcher, K., Robertson, L., Hora, J. L., Simons, D. A.,
Mays, W., Nolan, R., Bec, M., Merrill, M., & Fowler, A. M.
2003, PASP, 115, 1388
H?g, E., Fabricius, C., Makarov, V. V., Urban, S., Corbin, T.,
Wycoff, G., Bastian, U., Schwekendiek, P., & Wicenec, A. 2000,
A&A, 355, L27
Holman, M. J., & Wiegert, P. A. 1999, AJ, 117, 621
Houck, J. R., Roellig, T. L., van Cleve, J., Forrest, W. J., Herter,
T., Lawrence, C. R., Matthews, K., Reitsema, H. J., Soifer,
B. T., Watson, D. M., Weedman, D., Huisjen, M., Troeltzsch,
J., Barry, D. J., Bernard-Salas, J., Blacken, C. E., Brandl,
B. R., Charmandaris, V., Devost, D., Gull, G. E., Hall, P.,
Henderson, C. P., Higdon, S. J. U., Pirger, B. E., Schoenwald,
J., Sloan, G. C., Uchida, K. I., Appleton, P. N., Armus, L.,
Burgdorf, M. J., Fajardo-Acosta, S. B., Grillmair, C. J., Ingalls,
J. G., Morris, P. W., & Teplitz, H. I. 2004, ApJS, 154, 18
Hubeny, I., & Burrows, A. 2007, ApJ, 669, 1248
Janson, M., Bergfors, C., Goto, M., Brandner, W., & Lafreni`ere,
D. 2010, ApJ, 710, L35
Johnson, J. A., & Apps, K. 2009, ApJ, 699, 933
Jones, H. R. A., & Tsuji, T. 1997, ApJ, 480, L39+
Kalas, P., Graham, J. R., Chiang, E., Fitzgerald, M. P., Clampin,
M., Kite, E. S., Stapelfeldt, K., Marois, C., & Krist, J. 2008,
Science, 322, 1345
Kiraga, M., & Stepien, K. 2007, Acta Astronomica, 57, 149
Kirkpatrick, J. D. 2005, ARA&A, 43, 195
Kirkpatrick, J. D., Reid, I. N., Liebert, J., Cutri, R. M., Nelson,
B., Beichman, C. A., Dahn, C. C., Monet, D. G., Gizis, J. E.,
& Skrutskie, M. F. 1999, ApJ, 519, 802
Kirkpatrick, J. D., et al. 2008, ApJ, 689, 1295
Knapp, G. R., Leggett, S. K., Fan, X., Marley, M. S., Geballe,
T. R., Golimowski, D. A., Finkbeiner, D., Gunn, J. E.,
Hennawi, J., Ivezi?c, Z., Lupton, R. H., Schlegel, D. J., Strauss,
M. A., Tsvetanov, Z. I., Chiu, K., Hoversten, E. A.,
Glazebrook, K., Zheng, W., Hendrickson, M., Williams, C. C.,
Uomoto, A., Vrba, F. J., Henden, A. A., Luginbuhl, C. B.,
Guetter, H. H., Munn, J. A., Canzian, B., Schneider, D. P., &
Brinkmann, J. 2004, AJ, 127, 3553
Lafreni`ere, D., Jayawardhana, R., & van Kerkwijk, M. H. 2008,
ApJ, 689, L153
?. 2010, ApJ, 719, 497
Lagrange, A., Bonnefoy, M., Chauvin, G., Apai, D., Ehrenreich,
D., Boccaletti, A., Gratadour, D., Rouan, D., Mouillet, D.,
Lacour, S., & Kasper, M. 2010, Science, 329, 57
Lawrence, A., et al. 2007, MNRAS, 379, 1599
Leggett, S. K., Burningham, B., Saumon, D., Marley, M. S.,
Warren, S. J., Smart, R. L., Jones, H. R. A., Lucas, P. W.,
Pinfield, D. J., & Tamura, M. 2010, ApJ, 710, 1627
Leggett, S. K., Cushing, M. C., Saumon, D., Marley, M. S.,
Roellig, T. L., Warren, S. J., Burningham, B., Jones, H. R. A.,
Kirkpatrick, J. D., Lodieu, N., Lucas, P. W., Mainzer, A. K.,
Mart??n, E. L., McCaughrean, M. J., Pinfield, D. J., Sloan,
G. C., Smart, R. L., Tamura, M., & Van Cleve, J. 2009, ApJ,
695, 1517
Leggett, S. K., Marley, M. S., Freedman, R., Saumon, D., Liu,
M. C., Geballe, T. R., Golimowski, D. A., & Stephens, D. C.
2007, ApJ, 667, 537
Leggett, S. K., Saumon, D., Albert, L., Cushing, M. C., Liu,
M. C., Luhman, K. L., Marley, M. S., Kirkpatrick, J. D.,
Roellig, T. L., & Allers, K. N. 2008, ApJ, 682, 1256
Linsky, J. L. 1969, ApJ, 156, 989
Liu, F., Cutri, R., Greanias, G., Duval, V., Eisenhardt, P., Elwell,
J., Heinrichsen, I., Howard, J., Irace, W., Mainzer, A.,
Razzaghi, A., Royer, D., & Wright, E. L. 2008, in Society of
Photo-Optical Instrumentation Engineers (SPIE) Conference
Series, Vol. 7017, Society of Photo-Optical Instrumentation
Engineers (SPIE) Conference Series
Liu, M. C., Dupuy, T. J., & Leggett, S. K. 2010, ArXiv e-prints
Lodders, K. 1999, ApJ, 519, 793
?. 2002, ApJ, 577, 974
?. 2003, ApJ, 591, 1220
Lodders, K., & Fegley, B. 2002, Icarus, 155, 393
Lodders, K., & Fegley, Jr., B. 2006, Chemistry of Low Mass
Substellar Objects (Astrophysics Update 2), 1?+
Looper, D. L., Kirkpatrick, J. D., Cutri, R. M., Barman, T.,
Burgasser, A. J., Cushing, M. C., Roellig, T., McGovern,
M. R., McLean, I. S., Rice, E., Swift, B. J., & Schurr, S. D.
2008, ApJ, 686, 528
L?opez-Santiago, J., Micela, G., & Montes, D. 2009, A&A, 499, 129
Luhman, K. L., et al. 2007, ApJ, 654, 570
Madsen, S. 2003, A&A, 401, 565
Magazzu, A., Martin, E. L., & Rebolo, R. 1993, ApJ, 404, L17
Marchi, S. 2007, ApJ, 666, 475
Marley, M. 2000, in Astronomical Society of the Pacific
Conference Series, Vol. 212, From Giant Planets to Cool Stars,
ed. C. A. Griffith & M. S. Marley, 152?+
Marley, M. S., Saumon, D., & Goldblatt, C. submitted, ApJ
Marley, M. S., Seager, S., Saumon, D., Lodders, K., Ackerman,
A. S., Freedman, R. S., & Fan, X. 2002, ApJ, 568, 335
Marois, C., Macintosh, B., Barman, T., Zuckerman, B., Song, I.,
Patience, J., Lafreni`ere, D., & Doyon, R. 2008, Science, 322,
1348
Marsh, K. A., Kirkpatrick, J. D., & Plavchan, P. 2010, ApJ, 709,
L158
Martin, E. L., Rebolo, R., & Zapatero-Osorio, M. R. 1996, ApJ,
469, 706
McLean, I. S., McGovern, M. R., Burgasser, A. J., Kirkpatrick,
J. D., Prato, L., & Kim, S. S. 2003, ApJ, 596, 561
McLean, I. S., Prato, L., McGovern, M. R., Burgasser, A. J.,
Kirkpatrick, J. D., Rice, E. L., & Kim, S. S. 2007, ApJ, 658,
1217
McPherson, A. M., Born, A., Sutherland, W., Emerson, J., Little,
B., Jeffers, P., Stewart, M., Murray, J., & Ward, K. 2006, in
Society of Photo-Optical Instrumentation Engineers (SPIE)
Conference Series, Vol. 6267, Society of Photo-Optical
Instrumentation Engineers (SPIE) Conference Series
Metchev, S. A., & Hillenbrand, L. A. 2006, ApJ, 651, 1166
Mohanty, S., Jayawardhana, R., Hu?elamo, N., & Mamajek, E.
2007, ApJ, 657, 1064
Montes, D., L?opez-Santiago, J., Fern?andez-Figueroa, M. J., &
G?alvez, M. C. 2001, A&A, 379, 976
Moorwood, A., Cuby, J., Biereichel, P., Brynnel, J., Delabre, B.,
Devillard, N., van Dijsseldonk, A., Finger, G., Gemperlein, H.,
Gilmozzi, R., Herlin, T., Huster, G., Knudstrup, J., Lidman,
C., Lizon, J., Mehrgan, H., Meyer, M., Nicolini, G., Petr, M.,
Spyromilio, J., & Stegmeier, J. 1998, The Messenger, 94, 7
Nakajima, T., Oppenheimer, B. R., Kulkarni, S. R., Golimowski,
D. A., Matthews, K., & Durrance, S. T. 1995, Nature, 378, 463
Nelson, R. P. 2003, MNRAS, 345, 233
Nidever, D. L., Marcy, G. W., Butler, R. P., Fischer, D. A., &
Vogt, S. S. 2002, ApJS, 141, 503
Perryman, M. A. C., Brown, A. G. A., Lebreton, Y., Gomez, A.,
Turon, C., Cayrel de Strobel, G., Mermilliod, J. C., Robichon,
N., Kovalevsky, J., & Crifo, F. 1998, A&A, 331, 81
Pierens, A., & Nelson, R. P. 2008, A&A, 483, 633
Pinfield, D. J., Jones, H. R. A., Lucas, P. W., Kendall, T. R.,
Folkes, S. L., Day-Jones, A. C., Chappelle, R. J., & Steele, I. A.
2006, MNRAS, 368, 1281
Pizzolato, N., Maggio, A., Micela, G., Sciortino, S., & Ventura, P.
2003, A&A, 397, 147
Pollack, J. B., Hubickyj, O., Bodenheimer, P., Lissauer, J. J.,
Podolak, M., & Greenzweig, Y. 1996, Icarus, 124, 62
Reid, I. N., Hawley, S. L., & Gizis, J. E. 1995, AJ, 110, 1838
Saumon, D., Bergeron, P., Lunine, J. I., Hubbard, W. B., &
Burrows, A. 1994, ApJ, 424, 333
Saumon, D., & Marley, M. S. 2008, ApJ, 689, 1327
Saumon, D., Marley, M. S., Cushing, M. C., Leggett, S. K.,
Roellig, T. L., Lodders, K., & Freedman, R. S. 2006, ApJ, 647,
552
Saumon, D., Marley, M. S., Leggett, S. K., Geballe, T. R.,
Stephens, D., Golimowski, D. A., Cushing, M. C., Fan, X.,
Rayner, J. T., Lodders, K., & Freedman, R. S. 2007, ApJ, 656,
1136
Schlaufman, K. C., & Laughlin, G. 2010, ArXiv e-prints
Schmidt, S. J., West, A. A., Hawley, S. L., & Pineda, J. S. 2010,
AJ, 139, 1808
Schmidt, T. O. B., Neuh?auser, R., Seifahrt, A., Vogt, N.,
Bedalov, A., Helling, C., Witte, S., & Hauschildt, P. H. 2008,
A&A, 491, 311
20 Burgasser et al.
Scholz, R. 2010a, A&A, 515, A92+
?. 2010b, A&A, 510, L8+
Simcoe, R. A., Burgasser, A. J., Bernstein, R. A., Bigelow, B. C.,
Fishner, J., Forrest, W. J., McMurtry, C., Pipher, J. L.,
Schechter, P. L., & Smith, M. 2008, in Society of Photo-Optical
Instrumentation Engineers (SPIE) Conference Series, Vol. 7014,
Society of Photo-Optical Instrumentation Engineers (SPIE)
Conference Series
Simcoe, R. A., Burgasser, A. J., Bochanski, J. J., Schechter,
P. L., Smith, M. J., Fishner, J., Bernstein, R. A., Bigelow,
B. C., Pipher, J. L., Forrest, W. J., McMurtry, C., Blank, R.,
& Kumler, J. 2010, PASP
Simons, D. A., & Tokunaga, A. 2002, PASP, 114, 169
Soter, S. 2006, AJ, 132, 2513
Spiegel, D. S., Burrows, A., & Milsom, J. A. 2010, ArXiv e-prints
Stamatellos, D., & Whitworth, A. P. 2009, MNRAS, 392, 413
Stephens, D. C., Leggett, S. K., Cushing, M. C., Marley, M. S.,
Saumon, D., Geballe, T. R., Golimowski, D. A., Fan, X., &
Noll, K. S. 2009, ApJ, 702, 154
Takeda, G., Ford, E. B., Sills, A., Rasio, F. A., Fischer, D. A., &
Valenti, J. A. 2007, ApJS, 168, 297
Tinney, C. G., Burgasser, A. J., Kirkpatrick, J. D., & McElwain,
M. W. 2005, AJ, 130, 2326
Tody, D. 1986, in Society of Photo-Optical Instrumentation
Engineers (SPIE) Conference Series, Vol. 627, Society of
Photo-Optical Instrumentation Engineers (SPIE) Conference
Series, ed. D. L. Crawford, 733?+
Tokunaga, A. T., Simons, D. A., & Vacca, W. D. 2002, PASP,
114, 180
Tsuji, T. 2005, ApJ, 621, 1033
Tsuji, T., Ohnaka, K., & Aoki, W. 1999, ApJ, 520, L119
Vacca, W. D., Cushing, M. C., & Rayner, J. T. 2003, PASP, 115,
389
van Leeuwen, F. 2007, A&A, 474, 653
Veras, D., Crepp, J. R., & Ford, E. B. 2009, ApJ, 696, 1600
Visscher, C., Lodders, K., & Fegley, B. 2010, ApJ, 716, 1060
Voges, W., Aschenbach, B., Boller, T., Brauninger, H., Briel, U.,
Burkert, W., Dennerl, K., Englhauser, J., Gruber, R., Haberl,
F., Hartner, G., Hasinger, G., Pfeffermann, E., Pietsch, W.,
Predehl, P., Schmitt, J., Trumper, J., & Zimmermann, U. 2000,
IAU Circ., 7432, 3
Vorobyov, E. I., & Basu, S. 2010, ApJ, 714, L133
Vrba, F. J., Henden, A. A., Luginbuhl, C. B., Guetter, H. H.,
Munn, J. A., Canzian, B., Burgasser, A. J., Kirkpatrick, J. D.,
Fan, X., Geballe, T. R., Golimowski, D. A., Knapp, G. R.,
Leggett, S. K., Schneider, D. P., & Brinkmann, J. 2004, AJ,
127, 2948
Warren, S. J., et al. 2007, MNRAS, 381, 1400
West, A. A., Hawley, S. L., Bochanski, J. J., Covey, K. R., Reid,
I. N., Dhital, S., Hilton, E. J., & Masuda, M. 2008, AJ, 135, 785
Witte, S., Helling, C., & Hauschildt, P. H. 2009, A&A, 506, 1367
Wright, G. S., Reike, G., Barella, P., Boeker, T., Colina, L., van
Dishoeck, E., Driggers, P., Goodson, G., Greene, T., Heske, A.,
Henning, T., Lagage, P., Meixner, M., Norgaard-Nielsen, H.,
Olofsson, G., Ray, T., Ressler, M., Thatcher, J., Waelkens, C.,
Wright, D., & Zehnder, A. 2008, in Society of Photo-Optical
Instrumentation Engineers (SPIE) Conference Series, Vol. 7010,
Society of Photo-Optical Instrumentation Engineers (SPIE)
Conference Series
Yamamura, I., Tsuji, T., & Tanabe, T. 2010, ArXiv e-prints
Zapatero Osorio, M. R., B?ejar, V. J. S., Mart??n, E. L., Rebolo,
R., Barrado y Navascu?es, D., Mundt, R., Eisl?offel, J., &
Caballero, J. A. 2002, ApJ, 578, 536
Zapatero Osorio, M. R., Lane, B. F., Pavlenko, Y., Mart??n, E. L.,
Britton, M., & Kulkarni, S. R. 2004, ApJ, 615, 958