Exploring the substellar temperature regime down to ~550K
Ben Burningham, D. J. Pinfield, S. K. Leggett, M. Tamura, P. W. Lucas, D. Homeier, A. Day-Jones, H. R. A. Jones, J.R.A. Clarke, M. Ishii, M. Kuzuhara, N. Lodieu, M. R. Zapatero Osorio, B.P. Venemans, D.J. Mortlock, D. Barrado y Navascues, E. L. Martin, A. Magazzu
aa r X i v : . [ a s t r o - ph ] A ug Mon. Not. R. Astron. Soc. , 000–000 (0000) Printed 28 October 2018 (MN L A TEX style file v2.2)
Exploring the substellar temperature regime down to ∼ Ben Burningham ⋆ , D. J. Pinfield , S. K. Leggett , M. Tamura , P. W. Lucas ,D. Homeier , A. Day-Jones , H. R. A. Jones , J.R.A. Clarke ,M. Ishii , M. Kuzuhara , N. Lodieu , M. R. Zapatero Osorio , B.P. Venemans ,D.J. Mortlock , D. Barrado y Navascu´es ,E. L. Martin , A. Magazz`u Centre for Astrophysics Research, Science and Technology Research Institute, University of Hertfordshire, Hatfield AL10 9AB Gemini Observatory, 670 N. A’ohoku Place, Hilo, HI 96720, USA National Astronomical Observatory, Mitaka, Tokyo 181-8588 Institut fur Astrophysik, Georg-August-Universitat, Friedrich-Hund-Platz 1, 37077 Gottingen, Germany Subaru Telescope, 650 North A’ohoku Place, Hilo, Hi 96720, USA University of Tokyo, Hongo, Tokyo 113-0033, Japan Instituto de Astrof´ısica de Canarias, 38200 La Laguna, Spain Institute of Astronomy, Madingley Road, Cambridge CB3 0HA, UK Astrophysics Group, Imperial College London, Blackett Laboratory, Prince Consort Road, London SW7 2AZ Laboratorio de Astrof´ısica Espacial y F´ısica Fundamental,INTA, P.O. Box 78, E–28691 Villanueva de la Canada (Madrid), Spain Fundaci´on Galileo Galilei-INAF, Apartado 565, E-38700 Santa Cruz de La Palma, Spain
28 October 2018
ABSTRACT
We report the discovery of three very late T dwarfs in the UKIRT Infrared DeepSky Survey (UKIDSS) Third Data Release: ULAS J101721.40+011817.9 (ULAS1017),ULAS J123828.51+095351.3 (ULAS1238) and ULAS J133553.45+113005.2(ULAS1335). We detail optical and near-infrared photometry for all three sources, andmid-infrared photometry for ULAS1335. We use near-infrared spectra of each sourceto assign spectral types T8p (ULAS1017), T8.5 (ULAS1238) and T9 (ULAS1335)to these objects. ULAS1017 is classed as a peculiar T8 (T8p) due to appearing asa T8 dwarf in the J -band, whilst exhibiting H and K -band flux ratios consistentwith a T6 classification. Through comparison to BT-Settl model spectra we estimatethat ULAS1017 has 750K < ∼ T eff < ∼ . < ∼ log g ( cms − ) < ∼ .
5, assumingsolar metallicity. This estimate for gravity is degenerate with varying metallicity.We estimate that ULAS1017 has an age of 1.6–15 Gyr, a mass of 33–70 M J andlies at a distance of 31–54 pc. We do not estimate atmospheric parameters forULAS1238 due to a lack of K -band photometry. We extend the unified scheme ofBurgasser et al. (2006) to the type T9 and suggest the inclusion of the W J index toreplace the now saturated J -band indices. We propose ULAS1335 as the T9 spectraltype standard. ULAS1335 is the same spectral type as ULAS J003402.77-005206.7and CFBDS J005910.90-011401.3. We argue that given the similarity of the currentlyknown > T8 dwarfs to the rest of the T dwarf sequence, the suggestion of the Y0spectral class for these objects is premature. Comparison of model spectra withthat of ULAS1335 suggest a temperature below 600K, possibly combined withlow-gravity and/or high-metallicity. We find ULAS1335 to be extremely red innear to mid-infrared colours, with H − [4 .
49] = 4 . ± .
04 . This is the reddestnear to mid-infrared colour yet observed for a T dwarf. The near to mid-infraredspectral energy distribution of ULAS1335 further supports T eff < T eff ∼ − J and lies at a distance of 8–12 pc. Key words: surveys - stars: low-mass, brown dwarfs ⋆ E-mail: [email protected] (cid:13)
Ben Burningham et al.
Extending the known sample of field dwarfs to ever lowereffective temperatures ( T eff ) is important not only for thedetermination of the field mass function, but also for prob-ing temperature and pressure regimes that have hithertobeen unexplored observationally. The study of extremelycool brown dwarfs opens a window on atmospheric physicswhich will be fundamental for understanding the processeswithin a broad range of substellar atmospheres, includingthose of giant exoplanets.At the time of writing there are 142 T dwarfs publishedin the literature (e.g., DwarfArchives.org; Pinfield et al.2008). Of these, only five are classified as T8 or later(using the scheme of Burgasser et al. 2006), two of whichhave spectral types later than T8: ULAS J003402.77-005206.7 (hereafter ULAS0034; Warren et al. 2007)and CFBDS J005910.90-011401.3 (hereafter CFBDS0059;Delorme et al. 2008). ULAS0034 has an inferred T eff of 600-700 K (Warren et al. 2007), and Delorme et al. (2008) infera T eff for CFBDS0059 that is ∼
50 K cooler than that ofULAS0034.Expanding the sample of very cool brown dwarfs,and exploring possible factors that could motivate theimplementation of new spectral classes (e.g Y dwarfs;Kirkpatrick et al. 1999) are primary science drivers forthe UKIRT Infrared Deep Sky Survey (UKIDSS) LargeArea Survey (LAS; see Lawrence et al. 2007). Our col-laboration, the UKIDSS Cool Dwarf Science WorkingGroup (CDSWG) is engaged in a substantial effort toachieve this aim (e.g., Pinfield et al. 2008; Kendall et al.2007; Lodieu et al. 2007; Warren et al. 2007). Here wepresent the recent discovery by CDSWG of three verycool brown dwarfs. These are ULAS J101721.40+011817.9(ULAS1017), ULAS J123828.51+095351.3 (ULAS1238) andULAS J133553.45+113005.2 (ULAS1335). The last of these,ULAS1335, may be the coolest brown dwarf yet discoveredand we here discuss the properties of this object, estimatinga T eff of approximately 550–600 K. The objects presented here were selected from the third datarelease (DR3) of UKIDSS (Warren et al., in prep). We usedthe same method as that outlined in Sections 2 and 3 ofPinfield et al. (2008). The process of identifying and con-firming extremely late T dwarfs can be separated into threeprinciple stages: 1) initial selection by mining survey data;2) confirmation of photometric properties by near-infraredand optical follow-up; 3) near-infrared spectroscopy to esti-mate spectral type. The Gemini/NIRI spectroscopic follow-up took place in two stages in queue mode. Initial, short( t int = 16mins) J -band observations were obtained first.These initial spectra confirmed that two of our targets wereindeed of extremely late spectral type, and deeper J , H and K -band observations were then queued. Subaru/IRCS spec-troscopic follow-up took place separately for additional can-didates in classical mode and confirmed the very late type ofa third target. We will discuss each of these stages in moredetail in the following sections. We initially searched the UKIDSS LAS data for sources thatmet colour criteria: Y − J > . J − H < . σ < .
3, where σ is the uncertainty in the magnitude measured in eachof the three Y JH -bands. The resultant candidate list wascross-matched against the Sloan Digital Sky Survey (SDSS)Data Release 6 (DR6) using a pairing radius of 2 arcsecs.Objects with no SDSS counterpart, or whose SDSS-UKIDSS z ′ − J > .
0, were retained as potential candidates. We iden-tified some 50 candidate T dwarfs in this search. ULASJ1238and ULASJ1335 were picked out for immediate and rapidfollow-up due to survey colours that were reminiscent of theprototype > T8 dwarf ULAS J0034. ULAS1017 on the otherhand was selected due to its extremely blue J − H colourfrom UKIDSS photometry. Near infrared follow-up photometry was obtained usingthe UKIRT Fast Track Imager (UFTI; Roche et al. 2003)mounted on UKIRT, and the Long-slit Infrared ImagingSpectrograph (LIRIS; Manchado et al. 1998) mounted onthe William Herschel Telescope on La Palma. The mosaicswere produced using sets of jittered images, with individualexposure times, jitter patterns and number of repeats givenin Table 1. The data were dark subtracted, flatfield cor-rected, sky subtracted and mosaiced using ORAC-DR forthe UFTI data, and LIRIS-DR for the LIRIS data. We cali-brated our UFTI observations using UKIRT Faint Standards(Leggett et al. 2006), with a standard observed at a similarairmass for each target. All UFTI data were obtained underphotometric conditions, with seeing better than 0.9 arcsec.The wider field LIRIS data were obtained through thin cir-rus in variable seeing (0.8 - 2.0 arcsec) and were calibratedagainst 2MASS stars within the field of view. LIRIS usesa K s filter, and as such no transform was required for the2MASS stars. The K s magnitude for ULAS1017 was trans-formed to the MKO system using the transform derived byPinfield et al. (2008).We have also obtained optical z -band photometry usingthe ESO Multi-Mode Instrument (EMMI) mounted on theNew Technology Telescope at La Silla, Chile on the nights of2008 January 29 and 2008 January 30. These observationsare summarised in Table 1. For this optical follow-up we useda Bessel z -band filter (ESO Z z − band photometry using SDSS DR6sources present in our images as secondary standards. Wetransformed the Sloan z ′ ( AB ) magnitudes to the EMMI sys-tem using the transformation given by Warren et al. (2007)before using them to determine zero points, which had a typ-ical scatter of ± .
04 mags. The resulting EMMI photometryfor our targets was then transformed to the Sloan z ′ ( AB )system (see Warren et al. 2007). The results of our ground-based follow-up photometry are given in Table 2, which alsogives the original database photometry. c (cid:13) , 000–000 ery cool T dwarfs Object Filter Instrument UT Date Total integration time t int breakdownULAS1017 J UFTI 2008 Jan 17 300s (j = 5, r = 1, t exp = 60s) H UFTI 2008 Jan 17 2400s (j = 5, r = 6, t exp = 60s) K s LIRIS 2008 Mar 17 2400s (j = 5, r = 24, t exp = 20s) z EMMI
EMMI 2008 Jan 30 2400s (j = 1,r = 4, t exp = 600s)ULAS1238 J UFTI 2008 Jan 25 300s (j = 5, r = 1, t exp = 60s) H UFTI 2008 Jan 25 1800s (j = 5, r = 6, t exp = 60s) z EMMI
EMMI 2008 Jan 30 2400s (j = 1,r = 4, t exp = 600s)ULAS1335 Y UFTI 2008 Jan 16 540s (j = 9, r = 1, t exp = 60s) J UFTI 2008 Jan 16 300s (j = 5, r = 1, t exp = 60s) H UFTI 2008 Jan 16 900s (j = 5, r = 3, t exp = 60s) K UFTI 2008 Jan 16 900s (j = 5, r = 3, t exp = 60s) z EMMI
EMMI 2008 Jan 29 1200s (j = 1,r = 2, t exp = 600s) Table 1.
Summary of the observations obtained for near infrared and optical photometric follow-up. The breakdown of each integrationis given in the final column with the following notation: j = number of jitter points; r = number of repeats for jitter pattern; t exp =exposure time at each jitter point.Object Y J H K z ′ ( AB ) z ′ ( AB ) − J Y − J J − H H − K ULAS1017 19 . ± .
15 18 . ± .
07 19 . ± .
24 1 . ± . − . ± . . ± .
02 19 . ± .
02 19 . ± .
10 22 . ± .
09 3 . ± .
09 1 . ± . − . ± .
03 0 . ± . . ± .
07 18 . ± .
06 19 . ± .
15 0 . ± .
09 -0 . ± . . ± .
02 19 . ± .
02 22 . ± .
12 3 . ± .
12 0 . ± . − . ± . . ± .
05 17 . ± .
03 18 . ± .
12 18 . ± .
20 0 . ± .
06 -0 . ± .
13 -0 . ± . . ± .
04 17 . ± .
01 18 . ± .
01 18 . ± .
03 22 . ± .
10 4 . ± .
10 0 . ± .
04 -0 . ± .
01 -0 . ± . Table 2.
The available photometry for each of the new ultracool dwarfs. In each case the first row gives survey photometry from UKIDSSLAS (apermag3; see Dye et al. 2006). All three objects were undetected in SDSS DR6. The second row for each object gives the resultsof the follow-up photometry described in Section 2.2.
Spectroscopy in the
JHK -bands was obtained forULAS1238 and ULAS1335 using the Near InfraRed Im-ager and Spectrometer (NIRI; Hodapp et al. 2003) on theGemini North Telescope on Mauna Kea under programsGN-2007B-Q-26 and GN-2008A-Q-15. The InfraRed Cam-era and Spectrograph (IRCS; Kobayashi et al. 2000) onthe Subaru telescope on Mauna Kea was used to obtainthe JH spectrum for ULAS1017 and this was then fol-lowed with spectroscopy in the H and K -bands with Gem-ini/NIRI. Additionally, we obtained a Y -band spectrumfor ULAS1335 using the Infrared Spectrometer And ArrayCamera (ISAAC; Moorwood et al. 1998) mounted on VLTUT1 and ESO, Paranal in Director’s Discretionary time(program ID: 280.C-5067(A)).All observations were made up of a set of sub-exposuresin an ABBA jitter pattern to facilitate effective backgroundsubtraction, with a slit width of 1 arcsec. The length of theA-B jitter was 10 arcsecs. The observations are summarisedin Table 3. The NIRI observations were reduced using stan-dard IRAF Gemini packages. The Subaru IRCS JH spec-trum was also extracted using standard IRAF packages. TheAB pairs were subtracted using generic IRAF tools, and me-dian stacked. In the case of IRCS, the data were found to besufficiently uniform in the spatial axis for flat-fielding to beneglected. The ISAAC data were reduced and extracted us-ing the same steps, but implemented with the ESO ISAACpipeline version 5.7.0.For all three sets of data, a comparison argon arc framewas used to obtain a dispersion solution, which was then ap- plied to the pixel coordinates in the dispersion direction onthe images. The resulting wavelength-calibrated subtractedpairs had a low-level of residual sky emission removed by fit-ting and subtracting this emission with a set of polynomialfunctions fit to each pixel row perpendicular to the disper-sion direction, and considering pixel data on either side ofthe target spectrum only. The spectra were then extractedusing a linear aperture, and cosmic rays and bad pixels re-moved using a sigma-clipping algorithm.Telluric correction was achieved by dividing each ex-tracted target spectrum by that of an early A or F typestar observed just before or after the target and at a simi-lar airmass. Prior to division, hydrogen lines were removedfrom the standard star spectrum by interpolating the stellarcontinuum. Relative flux calibration was then achieved bymultiplying through by a blackbody spectrum of the appro-priate T eff . Data obtained for the same spectral regions ondifferent nights were co-added after relative flux calibration,each weighted by their exposure time.The spectra were then joined together using the mea-sured near-infrared photometry to place the spectra on anabsolute flux scale, and rebinned by a factor of three to in-crease the signal-to-noise, whilst avoiding under-samplingfor the spectral resolution. The IRCS JH spectrum ofULAS1017 was joined to the NIRI H and K -band spectrain a similar way, but the NIRI spectra were rebinned by afactor of three prior to joining. The resultant Y JHK and
JHK spectra for ULAS1017 and ULAS1335 are shown inFigure 1, where the combined spectra have been normalisedto unity at 1 . µ m and each offset for clarity. In the case of c (cid:13) , 000–000 Ben Burningham et al.
Object UT Date Integration time Instrument Spectral regionULAS1017 2008 Jan 25 12x300s IRCS JH K H ULAS1238 2007 Dec 21 4x240s NIRI J J H K K ULAS1335 2007 Dec 21 4x240s NIRI J J H K Y Table 3.
Summary of the near-infrared spectroscopic observations.
Figure 1.
The NIRI
JHK spectrum for ULAS1335 (bottom),the JH spectrum for ULAS1238 (middle). The IRCS J and NIRI HK spectrum of ULAS1017 is shown on the top row. ULAS1238, we have so far been unable to obtain a K -bandmagnitude, so only the JH spectrum is shown in Figure 1,the K -band spectrum is shown separately in Figure 2, nor-malised to unity at 2 . µ m. We obtained IRAC four-channel (3.55, 4.49, 5.73 and7.87 µ m) photometry of the source on 2008 March 5. Thedata were obtained as part of the Spitzer Space Telescope
Cycle 4 program × . ′′ × . ′′
2, yielding a5 . ′ × . ′ We used exposure times of 30 sand a 9-position medium-sized (52 pixels) dither pattern, for For more information about IRAC, seeFazio et al. (2004) and the IRAC Users Manual athttp://ssc.spitzer.caltech.edu/irac/descrip.html
Figure 2.
The NIRI K -band spectrum for ULAS1238. The errorspectrum is shown offset by -0.2. an exposure of 270s for each channel, and a total observingtime of 15.8 minutes to acquire all channels, slew, and settle.The data were reduced using the post-basic-calibrationdata mosaics generated by version 17.0.4 of the IRACpipeline. The mosaics were flat-fielded and flux-calibratedusing super-flats and global primary and secondary stan-dards observed by
Spitzer . We performed aperture photom-etry using an aperture with a 2-pixel (or 2 . ′′
4) radius, tomaximise the signal–to–noise ratio. The object was rela-tively isolated with the closest source ∼ of 1.205, 1.221, 1.363 and 1.571to channels 1 through 4, respectively. The photometry wasconverted from milli-Janskys to magnitudes on the Vega sys-tem using the zero-magnitude fluxes given in the IRAC DataHandbook (280.9, 179.7, 115.0, and 64.1 Jy for channels 1 to4, respectively). Photometric errors were derived from the Information about the IRAC pipeline and data products canbe found at http://ssc.spitzer.caltech.edu/irac/dh/c (cid:13) , 000–000 ery cool T dwarfs Band Magnitude (Vega) µ m3.55 15 . ± . . ± . . ± . . ± . Table 4.
Mid-infrared photometry for ULAS1335. Random un-certainties are quoted. An additional 3% uncertainty should beadded in quadrature to the the quoted random uncertainties toaccount for systematics. uncertainty images that are provided with the post-basic-calibration data. The magnitudes and errors are given inTable 4. Note that in addition to the errors in the table, thereare absolute calibration uncertainties of 2–3 per cent. Thereare also systematic uncertainties introduced by pipeline de-pendencies of comparable magnitude (Leggett et al. 2007b).We adopt the total photometric uncertainty to be the sumin quadrature of the values given in Table 4 plus 3%.
Two methods for establishing spectral type are used. Thefirst is by comparison of our spectra to the spectra of T spec-tral type standards as defined by Burgasser et al. (2006).Since the latest spectral type defined by that system isT8, for which the standard is 2MASS J04151954-0935066(hereafter 2MASS0415), we also compare our spectra to thetwo previously identified T8+ objects, ULAS0034 and CF-BDS0059.The Y and K band morphologies are more stronglydependent on metallicity and gravity than the J and H -bands. This is due to the influence of pressure dependentcollisionally induced H absorption in the K band (e.g.Burgasser et al. 2002), and the red side of the Y band fluxpeak (Allard et al. 2008). In addition, the blue side of the Y band peak is shaped by the red wing of the broad K I ab-sorption centered at 7700 ˚A, which is strongly dependent onmetallicity (Burrows et al. 2002). As such we base our spec-tral types on the comparisons of the J and H band spectraof our objects. In this case we normalise our J and H -bandspectra to unity at 1 . µ m and 1 . µ m respectively, to facil-itate comparison of their relative shapes, rather than heights(which will be discussed separately).The second method for establishing type is the use ofthe T dwarf spectral indices, calculated as flux ratios, whichare also defined by Burgasser et al. (2006). Aware that somespectral indices may become saturated for types later thanT8, we place our emphasis for determining type on the tem-plate comparison, and use this to assess the validity of thedefined spectral indices.Delorme et al. (2008) argue that ULAS0034 and CF-BDS0059 may represent the transition to Y spectral classdue to the onset of broad NH absorption in the blue side ofthe H -band flux peak. They put forward the NH – H ratioas a spectral index, and suggest that it traces the strength ofammonia absorption (see Table 5). However, given the ab-sence of calculated ammonia opacities in the near-infrared,and possible influence of nearby water absorption band we do not feel that this identification is certain. Pending fur-ther confirmation of this feature, we will refer to this indexas “NH ”– H . Whether the absorption feature attributed toammonia by Delorme et al. (2008) is indeed the first defin-ing feature of the Y class will depend on how this featuredevelops towards lower temperatures and as more objects inthis regime are identified.We calculate the W J and “NH ”– H spectral indicesas put forward by Warren et al. (2007) and Delorme et al.(2008) respectively. The W J index traces the decreasingwidth of the J -band peak through the ratio of flux in asection of the blue slope of the peak to that in the crown.As such the value W J index is driven by the strength ofthe J -band water absorption. These indices and the flux ra-tios from which they are derived are given in Table 5 for thethree objects we have introduced, in addition to those for theT8 standard 2MASS0415 and the two previously discoveredT8+ objects. The final two columns of Table 5 give the im-plied type from the template comparison and the spectraltype we adopt for each object, as discussed in more detailin the following sections. The spectrum of ULAS1017 has a peculiar property. It ap-pears to have a different spectral type in the H and K -band flux peaks as compared to the J -band peak. Fig-ure 3 shows the comparison of our IRCS+NIRI spectrumfor ULAS1017 with the T8 standard 2MASS0415 and the T6standard SDSS J162414.37+002915.6 (hereafter SDSS1624).ULAS1017 traces the form of the T8 spectrum well in the J -band, but the H -band peak matches the T6 spectrum best,particularly towards the the red end of this region. This isreflected in the calculated spectral indices for this object (seeTable 5). We adopt therefore the type T8p for this object.We do not think such morphology could arise as a re-sult of a T8 + T6, or similar, unresolved binary, as fluxfrom the brighter T6 would tend to fill-in the deep absorp-tion feature in the J -band, where we see very little flux inthis case. Furthermore, spectral synthesis using various bi-nary components, following the method of Burgasser et al.(2008), is not able to reproduce the spectral ratios seen forthis object. The problem arises because earlier type objects,which can reproduce the H and K bands tend to dominatethe light, and drive the ratios in the J band to earlier val-ues also. Although J -band flux reversals have been observedelsewhere (e.g., Looper et al. 2008), the late spectral typeof this object would seem to preclude that interpretation.Pinfield et al. (2008) identified ULAS J1150+0949, a T6.5pwith a similarly peculiar early H -band type (of T3 in thiscase), so such morphology may be a generic trait for someobjects. c (cid:13) , 000–000 Ben Burningham et al.
Figure 3.
The IRCS+NIRI
JHK spectrum of ULAS1017 (black line) overplotted with the spectra of the T8 standard 2MASS0415 andthe T6 standard SDSS1624. The error spectrum, offset by -0.2 in the y-axis, is plotted as a black line.c (cid:13) , 000–000 ery cool T dwarfs Figure 4.
The J and H -band flux peaks of ULAS1238 (black line), ULAS0034 (T9, see text; green line) and 2MASS0415 (T8; red line).The black line that is plotted offset by -0.2 in the y-axis is the error spectrum for ULAS1238. Also indicated are the spectral regionsintegrated to form the numerator of the spectral indices discussed in the text. The region that forms the denominator is indicated withthe letter “D”. Figure 5.
The J and H -band flux peaks of ULAS1335 (black line), ULAS0034 (T9, see text; green line) and CFBDS0059 (T9, see text;red line). The black line that is plotted offset by -0.2 in the y-axis is the error spectrum for ULAS1335. Also indicated are the spectralregions integrated to form the numerator of the spectral indices discussed in the text. The region that forms the denominator is indicatedwith the letter “D”.c (cid:13) , 000–000 B e n B u r n i n g ha m e t a l . Index H2O - J CH4 -
J WJ
H2O - H CH4 - H “NH3” - H CH4 -
K K / J Template TypeFlux ratio R . . f ( λ ) dλ R . . f ( λ ) dλ R . . f ( λ ) dλ R . . f ( λ ) dλ R . . f ( λ ) dλ R . . f ( λ ) dλ R . . f ( λ ) dλ R . . f ( λ ) dλ R . . f ( λ ) dλ R . . f ( λ ) dλ R . . f ( λ ) dλ R . . f ( λ ) dλ R . . f ( λ ) dλ R . . f ( λ ) dλ R . . f ( λ ) dλ R . . f ( λ ) dλ . ± .
003 0.050 ( > T7) 0.134 ... T8ULAS0034 0 . ± .
006 ( > T8) 0 . ± .
009 ( > T8) 0 . ± .
006 0 . ± .
010 ( > T8) 0 . ± .
006 (T8) 0 . ± .
008 0 . ± .
015 ( > T7) 0 . ± . > T8 T9CFBDS0059 0 . ± .
005 (T8) 0 . ± .
008 (T8) 0 . ± .
004 0 . ± .
008 ( > T8) 0 . ± .
002 (T8) 0 . ± .
005 0 . ± .
037 ( > T7) 0 . ± . > T8 T9ULAS1017 0 . ± .
007 (T8) 0 . ± .
006 (T8) 0 . ± .
005 0 . ± .
016 (T6) 0 . ± .
011 (T6) 0 . ± .
017 0 . ± .
018 (T6) 0 . ± .
002 T8/T6 T8pULAS1238 0 . ± .
008 (T8) 0 . ± .
009 (T7) 0 . ± .
006 0 . ± .
013 ( > T8) 0 . ± .
009 (T8) 0 . ± .
013 0 . ± .
012 ( > T7) ... T8 T8.5ULAS1335 0 . ± .
004 (T8) 0 . ± .
004 (T8) 0 . ± .
003 0 . ± .
004 ( > T8) 0 . ± .
003 (T8) 0 . ± .
002 0 . ± .
007 ( > T7) 0 . ± . > T8 T9
Table 5.
The near-infrared spectral index ratios as defined by Burgasser et al. (2006) for the three objects presented in this paper, and for the T8 standard 2MASS0415 and thepreviously discovered T8+ dwarfs. Also given is the W J index as defined by Warren et al. (2007), and the “NH ”- H index defined by Delorme et al. (2008). The final two columns givethe spectral type from template comparison in the Burgasser et al. (2006) scheme, and the final adopted spectral type. c (cid:13) R A S , M N R A S , ery cool T dwarfs Figure 6.
The full
Y JHK comparison of ULAS1335 toULAS0034, CFBDS0059 and 2MASS0415.
Figure 4 compares the J and H -band flux peaks forULAS1238 to those for 2MASS0415 and ULAS0034. Initialexamination of these plots, and the spectral indices given inTable 5, suggest a spectral type of T8 for ULAS1238. Closerexamination of the H -band peak reveals some similar nar-row features in the spectra of ULAS0034 and ULAS1238.Additionally, the width of the J -band peak, and the valuesof the H -band spectral indices imply that ULAS1238 is infact intermediate between ULAS0034 and 2MASS0415. Ifwe adopt the type T9 for ULAS0034 (see Section 3.3), thenULAS1238 could reasonably be classified as T8.5. However,the inherently small changes in near-infrared spectral mor-phology in this regime make such distinctions non-critical inthe absence of additional mid-infrared spectral coverage. An initial inspection of the shapes of the J and H -bandflux peaks of ULAS1335 (see Figure 5) indicates that it isof the same spectral type as ULAS0034, and CFBDS0059.In particular it should be noted that the strength of the ab-sorption blueward of 1 . µ m that Delorme et al. (2008)attribute to NH is similar for all three objects. Given theclose similarity of these objects to the rest of the T dwarfsequence, it is appropriate that these objects form an ex-tension of that sequence, and we assign to them the sub-type T9. Figure 6 shows the full Y JHK comparison ofULAS1335, CFBDS0059, ULAS0034 and 2MASS0415. Al-though the shapes of the H -band peaks are similar for thethree objects, their height relative to the J -band peak variessignificantly. Specifically, ULAS1335 and CFBDS0059 bothexhibit much stronger H -band peaks than either ULAS0034or 2MASS0415.In Figure 7 we plot the J and H band spectral indicesagainst the spectral types derived by template comparisonfor objects with types T6 to T9 (see below). It is apparentthat the water and methane J -band indices previously usedfor the classifying T dwarfs are degenerate with type for thesubtypes T8 and T9. Examination of Figure 7 indicates that the same may be true for the CH – H index, although theH O– H index is still able to distinguish these objects. Thenew indices “NH ”– H and W J both appear able to distin-guish the T9 dwarfs from the earlier type objects, howeverthe “NH ”– H index is degenerate with type for the earlierobjects.In Figure 8 we plot the “NH ”– H index against the W J spectral index suggested by Warren et al. (2007). It is ap-parent that the latest objects are well separated from T8objects in this plot. However, the values of W J and “NH ”– H do not show a consistent correlation for the latest objects.Figure 9 shows the H O– H index plotted with W J . Again,the three latest objects are well separated from the otherobjects, however in this case the values of the two indicesshow a clear trend (with the exception of the T8p dwarf),both decreasing together. The scatter in “NH ”– H valuesin Figure 8, compared to the correlation shown in Figure 9thus suggests that although “NH ”– H is useful for identify-ing T8+ objects, it is not an effective typing index for verylate objects in the range we are considering, and does notconveniently strap on to the previous T dwarf sequence.We suggest that the appropriate spectral indices for theextension of the T-spectral class to T9 are H O– H and W J .To provide continuity with the earlier T-indices we define therange of W J and H O– H for types T7-T9, matching our val-ues to the H O– H values of Burgasser et al. (2006) for typesT7 and T8, and using the apparent correlation in Figure 9 toarrive at appropriate W J values. Our proposed set of spec-tral indices for the latter portion of the T spectral sequenceare summarised in Table 6. In this scheme, ULAS1335,as well as ULAS0034 and CFBDS0059, are all T9 dwarfs,which are our adopted types. Since ULAS1335 is signifi-cantly brighter than both ULAS0034 and CFBDS0059, wealso propose ULAS1335 as the template for the T9 subtype.In Section 5.2 we estimate a T eff for ULAS1335 which ismarginally cooler than the Warren et al. (2007) estimate forULAS0034. This temperature sequence is well representedby the locations of the objects in Figure 9, but not by thesequence shown in Figure 8. If the “NH ”– H index is indeedtracing a broad ammonia absorption band, this suggests twothings. Firstly, it appears to develop at odds with the waterabsorption in the J -band (which the W J feature traces) andthe H O– H absorption feature, Secondly, if the relative tem-peratures of ULAS0034 and ULAS1335 are to be believed,it does not correlate well with temperature.There is some theoretical justification for expectingnear-infrared ammonia absorption to be only weakly depen-dent on T eff . Ammonia chemistry is governed by upmixing,and in practice the mixing ratio depends most strongly onthe ammonia abundance at the quench level. Since this isat a nearly fixed temperature, ∼ T eff . Giventhis behaviour, there is exciting potential for the “NH ”– H index as an additional diagnostic tool for very cool T dwarfs,and its development will be followed with interest as moremore objects in the very low-temperature regime are iden-tified. c (cid:13) , 000–000 Ben Burningham et al.
Figure 7.
Spectral index versus T-subtype for T6-T9 dwarfs. Asterisks indicate “normal” dwarfs, whilst the diamonds indicate two T8pdwarfs: ULAS1017 and 2MASS J07290002-3954043. Index values for these objects are drawn from Burgasser et al. (2006),Warren et al.(2007), Delorme et al. (2008) and Looper et al. (2007), or are calculated from the objects’ spectra supplied by these authors. Uncertaintiesare smaller than the symbol sizes.NIR spectral type H O-J CH -J W J H O- H CH - H CH -K StandardT7 0.07–0.13 0.21–0.28 0.35–0.40 0.20–0.26 0.15–0.25 < < < < < . < .
14 ... ... ULAS1335
Table 6.
Our proposed spectral indices for the latter-most portion of the T spectral sequence. Also given are the spectral type standardsfor each type.
It has been known for some time that the degree of multi-plicity amongst very young stars is greater than that of themore evolved field star populations (Duquennoy & Mayor1991; Leinert et al. 1993), and thus that the majority of bi-nary systems form together in their nascent clouds. Binarycomponents can therefore generally be assumed to share thesame age and composition, and wide common proper motioncompanions to our new objects could thus provide valuableconstraints if they are to be found. In order to address thispossibility and more generally assess the kinematics of ourobjects, we calculated proper motions for our targets usingthe highest signal-to-noise multi-epoch observations avail-able. In each case this was the WFCAM and UFTI J -bandimages.The IRAF routines geomap and geoxytran were used totransform between the available multi-epoch images, usingan average of 12 reference stars. This allowed any motion of the T dwarfs to be accurately measured. Proper motion un-certainties were estimated from centroiding accuracies com-bined with the residuals associated with our derived trans-formations. The calculated proper motions are given in Ta-ble 7. We also estimate V tan based on our distance estimatesderived in Section 7. These estimates lie within the range of V tan observed by Vrba et al. (2004) for L and T dwarfs.We independently searched the Supercosmos Sky Sur-vey database, and the Simbad database (accessing the Hip-parcos and Tycho catalogues amongst others), for commonproper motion companions (within 1 σ ) of the three new Tdwarfs, thus searching for wide binary companions over awide range of mass. The search was made within a projecteddistance of 20,000 AU from each target, calculated assumingthe minimal distance within our derived constraints (see Sec-tion 7) to ensure a spatially complete search. In the case ofULAS1238 we used a lower distance estimate of 20 pc basedon comparison with ULAS0034. SDSS and UKIDSS pho-tometry were then retrieved for any common proper motion c (cid:13) , 000–000 ery cool T dwarfs Figure 8.
The “NH ”– H index versus W J index. Uncertaintiesare not displayed as they are smaller than the symbol size. Figure 9.
The H O– H index versus W J index. Uncertainties arenot displayed as they are smaller than the symbol size. objects, and from these a spectral type range was estimated,and an approximate distance modulus inferred. Sources withdistance moduli that were inconsistent with the inferred dis-tances to our newly discovered dwarfs (see Section 7) couldthen be rejected as potential companions. Our search pro-duced no potential companions for any of our targets. In the following two sections we estimate the atmosphericparameters of ULAS1017 and ULAS1335 through modelcomparisons, and, in the case of ULAS1335, considerationof the near-infrared to mid-infrared spectral energy distri-bution (SED). The model spectra we employ are a recentrealisation of the BT-Settl models, for which detailed de-scriptions of the input physics are given in Warren et al.(2007) and Allard et al. (2008). There are uncertainties inthese models that limit their use significantly for simplisticcomparisons, but by limiting our analysis to relative com-parisons over key spectral ranges, and some extrapolationof theoretical trends, we are able to derive a set of “best-guess” properties based on the existing theoretical and ob-servational data.We do not perform a detailed analysis for ULAS1238due to the absence of K -band photometry that precludesflux calibration for this part of the spectrum. Instead wespeculate that its properties are likely intermediate between2MASS0415 and ULAS0034 since its spectral type is alsointermediate. However, we do not rule out the possibility ofa considerably lower T eff (see Section 6). To estimate the atmospheric parameters for ULAS1017 werepeat the ( W J , K/J ) analysis described by Warren et al.(2007) for ULAS0034, and applied to CFBDS0059 byDelorme et al. (2008). To this end we have calculated W J and K/J indices for a grid of solar metallicity BT-Settl mod-els with 4 . < log g < .
50, and 600
K < T eff < K .We have anchored the resulting ( W J , K/J ) grid such that T eff = 750 K and log g = 5.0 for the empirical ( W J , K/J ) co-ordinates of 2MASS0415, consistent with the detailed anal-ysis of Saumon et al. (2007). Placing ULAS1017 on this gridimplies a T eff = 800K and log g =5.5.In Figure 11 we compare the spectrum of ULAS1017to model spectra that bracket its location in the ( W J , K/J )plane. It appears that the model spectra generally fit thespectrum of ULAS1017 well. We note, however, that it isthe model for T eff = 800K and log g = 5.0 that provides thebest fitting spectrum, in contrast to the ( W J , K/J ) analysisthat suggests log g = 5.5.The location of ULAS1017 in Figure 10 is veryclose to that of Gl 570D. It is worth noting that us-ing detailed analysis with consideration of non-equilibriumeffects,Saumon et al. (2006) derive parameters for Gl 570Dof T eff = 800 − g = 5.09-5.23. This is verysimilar to our best value from spectral comparison forULAS1017, but similarly at odds with the parameters fromsimple ( W J , K/J ) analysis.Note also that this form of analysis can only provide ap-proximate estimates of the atmospheric parameters, and itmust be remembered that varying estimates of surface grav-ity will be degenerate with respect to varying metallicity.The lack of Y -band coverage in our spectrum of ULAS1017prevents us from making any estimate of its metallicity rel-ative to other objects based on the form of its Y -band peak.Furthermore, the form of the H -band flux peak, resemblingthat of a T6 dwarf adds an additional element of uncertaintyto our “best-guess” properties for this object. c (cid:13) , 000–000 Ben Burningham et al.
Object 1st Epoch Coordinates Epoch 1 Epoch 2 µ αcosδ µ δ V tan α δ mas yr − mas yr − − ULAS1017 10 17 21.40 +01 18 17.9 2007-03-17 2008-01-17 110 ±
80 40 ±
80 4-58ULAS1238 12 38 28.51 +09 53 51.3 2007-02-13 2008-01-25 418 ±
40 21 ±
40 ...ULAS1335 13 35 53.45 +11 30 05.2 2007-04-21 2008-01-16 160 ± − ±
50 10-21
Table 7.
Proper motion estimates for the three new T dwarfs. Uncertainties reflect the residuals in the transformations between the twoimages in each case. Also given are estimates for V tan for ULAS1335 and ULAS1017 (see Section 7 for distances). Figure 10. W J versus K/J indices for a grid of solar metallicityBT-Settl model spectra. The grid has been normalised by match-ing the T eff = 750 K , log g = 5.0 model indices to the empiricalindex values for 2MASS0415, which lies at (1.0, 1.0). We will begin by placing ULAS1335 on the ( W J , K/J ) grid(see Figure 10), which implies T eff ≈ g ≈ . W J , K/J ) plane for solar metal-licity. It is clear that these synthetic spectra are not a goodfit to the data. The most obvious discrepancy is the heightof the H -band flux peak. It is widely acknowledged that theshape of the H -band peak is poorly reproduced by modelspectra, a fact that is attributed to incomplete methane linelists. However, in the cases of ULAS0034, CFBDS0059 and2MASS0415, the observed height of the H -band flux peak iswell represented by the model spectra for the atmosphericparameters estimated from their location in the ( W J , K/J )plot (see Warren et al. 2007; Delorme et al. 2008, figures 6and 7 respectively) It is thus clear that despite many sim-ilarities between ULAS1335 and the other two T9 dwarfs,its H -band peak is, by comparison, significantly brighter.Some of the properties of ULAS1335 are thus presumablysomewhat different to the other two T9 dwarfs. Figure 11.
A comparison of model spectra for the parame-ters bracketing the location of ULAS1017 in the solar metallicity( W J , K/J ) plane.
We can make a first assessment of possible differencesin metallicity by considering the shape of the Y -band peak.All available theoretical models indicate that the shape ofthis flux peak will be sensitive to [M/H], although there issome ambiguity as to exactly how compositional changeswill affect its shape. A comparison of the Y -band peaks ofthe three T9 dwarfs in Figure 6 reveals them to have anextremely similar Y -band spectral morphology, and it thusseems unlikely that these objects have significantly differentmetallicity.Figure 12 shows what changes we might expect for vary-ing log g and T eff . In the bottom plot it can be seen thatalthough a higher gravity can increase the brightness of the H -band peak relative to the J − and Y -band peaks, it has c (cid:13)000
We can make a first assessment of possible differencesin metallicity by considering the shape of the Y -band peak.All available theoretical models indicate that the shape ofthis flux peak will be sensitive to [M/H], although there issome ambiguity as to exactly how compositional changeswill affect its shape. A comparison of the Y -band peaks ofthe three T9 dwarfs in Figure 6 reveals them to have anextremely similar Y -band spectral morphology, and it thusseems unlikely that these objects have significantly differentmetallicity.Figure 12 shows what changes we might expect for vary-ing log g and T eff . In the bottom plot it can be seen thatalthough a higher gravity can increase the brightness of the H -band peak relative to the J − and Y -band peaks, it has c (cid:13)000 , 000–000 ery cool T dwarfs the opposite effect on the K -band flux peak. These trendsthus indicate that it is not possible to theoretically accountfor the relatively bright H -band peak of ULAS1335 com-pared to the other T9 dwarfs, simply by varying log g alone.We can obtain a better match between theoretical andobserved spectra if we adopt a lower T eff for ULAS1335. Thetop plot in Figure 12 demonstrates the trend in the BT-Settl model spectra, and shows that the relative brightnessof the H -band peak increases compared to the Y - and J -band peaks as T eff decreases below 700K. However, notethat although there is some change to the relative strengthof the K -band peak, it is not as drastic as when one varieslog g . The overall near infrared morphology of ULAS1335 isthus best explained (through current theory) if this objecthas a somewhat lower T eff than ULAS0034.In Figure 13 we compare the spectra of ULAS1335to a range of model spectra with different metallicitiesand gravities for our lowest available temperature models, T eff = 600K. In the top plot, the combination log g = 5.0, T eff = 600K model spectra all estimate the height of the H -band peak well. The sub-solar metallicity model is a poormatch everywhere else however. The high-metallicity modelgives a reasonable approximation to the observed spectrum,but as expected it under-estimates the K -band flux some-what, and has a significantly wider and blue-ward shifted Y -band peak.In the bottom plot, the log g =4.25-4.50 model spectraprovide a reasonable fit to the whole near infrared spectra.However, whilst reproducing the Y -, J -, and K -band peaks,these spectra still somewhat under-estimate the brightnessof the H -band peak. Although the low T eff of this modelhas gone some way to addressing the relative brightness ofthis peak, the model trends suggest that we should obtain abetter fit by going to even lower T eff .Investigation of the near-infrared to mid-infrared SEDfor ULAS1335 reveals that is extremely red. Figure 14 showsthe 1-8 µ m SEDs for ULAS1335, ULAS0034, 2MASS0415and Gl 570D. Whilst the form of the SEDs are simi-lar, ULAS1335 is significantly redder than ULAS0034 overthe 2 - 4.49 µ m range. ULAS1335 has H − [4 .
49] =4 . ± .
04, the reddest value yet observed for a T dwarf.Warren et al. (2007) identified a correlation between the H –[4.49] colour and T eff for T dwarfs with well determined T eff (Golimowski et al. 2004). In fact the correlation for Tdwarfs with T eff = 800 − ± H –[4.49]colours of 2MASS0415, ULAS0034 and ULAS1335. Theo-retical tracks are also over-plotted for a recent realisationof the Marley et al. (2002) model atmospheres for the range500–800K (Marley & Saumon 2008). Tracks are shown for4 . log g .
48 and − . [ M/H ] +0 .
3, bracketingthe expected properties of ULAS1335. We have anchored themodels for log g = 5 . H -[4.49] and T eff for such log g and [M/H] ranges, and thisextrapolation suggests that ULAS1335 has T eff ∼ W J , K/J ) analysis was reasonably consistent with that in-ferred from an extrapolation of the T eff – H –[4.49] relation. Figure 12.
A comparison of model spectra for parameters brack-eting the location of ULAS1335 in the solar metallicity ( W J , K/J )plane.
However, in the case of ULAS1335 we have seen that thetwo estimates differ by ∼ W J , K/J ) method. It is likely that this disagree-ment is principally due to the fact that T9 dwarfs are in thetemperature range near the limit of where the near-infraredmodel spectra are valid. Since no model spectra cooler than600 K are plotted, a degeneracy in ( W J , K/J ) for these tem-peratures cannot be ruled out. Indeed, Figure 10 indicatesthat there is a predicted distortion in the grid (which willresult in some degeneracy) for the lowest gravity and tem-perature combinations currently considered.The strong, relatively tight T eff correlation offered byobserved T dwarf H –[4.49] colours combined with the the-oretical expectations for a reasonably steady shift of fluxfrom the near-infrared into the mid-infrared with decreas-ing T eff , lends weight to this colour as a better temperatureindicator for such very late objects, and we adopt the tem-perature implied by this colour rather than that implied bythe ( W J , K/J ) method. We also suggest that the ( W J , K/J )method be used with caution in the T eff regime near thelimit of current atmospheric models.To summarise, our comparison to model near infraredspectra suggested that ULAS1335 has T eff < c (cid:13) , 000–000 Ben Burningham et al.
Figure 13.
A comparison of model spectra with ULAS1335 fordiffering gravity and metallicity combinations at T eff = 600K. is clearly supported by the the near-to-mid infrared SED,which suggests T eff ∼ − T eff ∼ H band peak with decreasing T eff . In addition theheight of the H band peak appears more strongly depen-dent on metallicity than in the BT-Settl models, increasingwith decreasing metallicity.Finally, considering the absence of mid-infrared pho-tometry for CFBDS0059, and the reliance on the ( W J , K/J )method for its temperature estimate, we also note thatno strong conclusion regarding its temperature relative toULAS1335 and ULAS0034 should be drawn. We cannot ruleout the possibility that CFBDS0059 is cooler still.
Figure 14.
The 1-8 µ m SEDs for ULAS1335,ULAS0034(Warren et al. 2007), 2MASS0415 and Gl 570D. The photome-try for the latter two objects, taken from Leggett et al. (2002)and Patten et al. (2006), has been offset by +2.5 magnitudes toallow comparison with the T8+ dwarfs. Uncertainties are of com-parable size to the symbols. Figure 15.
The T eff - H –[4.49] relation for T dwarfs as definedby Warren et al. (2007). The shaded regions indicate the colours( ± σ ) for the coolest T dwarfs with IRAC magnitudes.Thestraight red line indicates the empirical fit the colours of warmerT dwarfs. Also plotted, as black lines are theoretical tracks fromMarley et al. (2002) and Marley & Saumon (2008). Values forlog g and metallicity are indicated in brackets.c (cid:13) , 000–000 ery cool T dwarfs We have also examined our T dwarf spectra in more detail,to try and identify new features that might be useful in thisnew T eff regime. We follow Leggett et al. (2007a) and use thespectrum of Jupiter along with transmission spectra of am-monia and methane as a basis for comparison, which we plotin Figure 16 alongside ULAS0034, CFBDS0059,ULAS1238and ULAS1335. Although the metallicity of the Jovian at-mosphere is much higher, and the gravity much lower, thanthat expected for our T dwarfs, this comparison with a muchcooler atmosphere is still interesting.We draw attention to the narrow feature near 1.23 µ m(left-most vertical green line in Figure 16) which appearsto be shared by Jupiter, ULAS1238 and ULAS1335. Thisshould not be confused with the K I doublet at ∼ . − . µ m. We attribute this feature to methane, based oncomparison with the laboratory transmission spectrum ofCruikshank & Binder (1969). We also highlight the narrowfeature at ∼ . µ m (right-most vertical green line), whichalso appears to be common between ULAS1238, ULAS1335and Jupiter. This may be due to ammonia, based on com-parison to the transmission spectrum of Irwin et al. (1999).We note, however that the sharp feature blueward of this(indicated by a vertical red line); also likely due to ammo-nia is apparently absent from the spectra of the T dwarfs. Since metallicity is thought to only weakly impact thederivation of fundamental parameters from T eff and log g (e.g., Saumon et al. 2007), we use the solar metallicity rela-tions summarised by Burrows et al. (2001) to estimate themass, age and radius for ULAS1017 and ULAS1335 fromour T eff and log g estimates.For ULAS1017, the extreme cases from the model com-parison, T eff = 750 − g = 5.0–5.5 correspond toan age of 1.6–15 Gyr and a mass of 33–70 M J . By estimatingits radius from the same set of relations, and normalising thecorresponding atmospheric models to the J -band flux peak,we can estimate the distance to ULAS1017 to be in the range31-54pc.For ULAS1335, we use our best estimate of T eff = 550 − g = 4 . − . J .In the absence of model spectra for T eff < T eff = 600K and log g = 4.5, scaling the syn-thetic parts with our z band photometry in the optical andour IRAC [4.49] photometry in the mid-infrared. We thenintegrated this this pseudo-synthetic spectrum over all wave-lengths to estimate the bolometric flux from ULAS1335. Theuncertainty in this estimate is dominated by uncertainty inthe photometry used to calibrate the spectrum. These un-certainties range from 10% for the optical photometry to <
5% for our near-infrared photometry. These regions con-tribute ∼ .
5% and ∼
25% to the bolometric flux estimaterespectively. The mid-infrared, which contributes ∼
75% ofthe bolometric flux, has an uncertainty of 4% in its [4.49]magnitude, which we use to scale the synthesised spectrum.Allowing for additional uncertainty in the model spectrummatch to the true mid-infrared spectrum, we estimate a to-tal uncertainty in our bolometric flux to be ∼ F bol = 3 . ± . × − Wm − .The uncertainty in our distance estimate for ULAS1335is dominated by the uncertainty in T eff (and thus luminos-ity). For T eff = 550 − g = 4.5–5.0 we estimatea luminosity of − . log ( L ∗ /L ⊙ ) − .
86 (by in-ferring radii using the relations in Burrows et al. 2001) andthus a distance of 8–12 pc. Table 8 summarises the inferredproperties for the T dwarfs identified in this paper.
We have identified three very late-type T dwarfs in theUKIDSS LAS DR3: ULAS1017, ULAS1238 and ULAS1335,for which we have adopted the spectral types T8p, T8.5 andT9 respectively. ULAS1017 has a peculiar spectrum, with a J -band typical of a T8 dwarf, but H and K -band peaks ofa T6. In assigning spectral types to the our targets, we havedefined the extension of the T-dwarf sequence to the typeT9, using the H O– H and W J indices, and ULAS1335 asthe spectral standard.To estimate atmospheric parameters for ULAS1017 andULAS1335, we have performed a detailed comparison withBT-Settl model spectra. For ULAS1017, we have estimatedthat its temperature lies in the range 750 T eff g = 5 . − . T eff that is cooler than the coolest available BT-Settl mod-els ( T eff < H − [4 .
49] = 4 . ± .
04. By extrapolating the em-pirical T eff – H –[4.49] relation to lower T eff , and guidedby model trends, we have estimated that ULAS1335 has T eff ∼ − ACKNOWLEDGMENTS
We wish to thank our anonymous referee for helpful sugges-tions that have improved the quality of this paper. We alsothank P. Delorme for supplying us with an electronic ver-sion of the spectrum for CFBDS0059 for use in this work.We are extremely grateful to D. Saumon and M. Marley forproviding us with model H -[4.49] colours for their latest setof ultracool dwarf atmospheres. The UKIDSS project is de-fined in Lawrence et al. (2007). UKIDSS uses the UKIRTWide Field Camera (WFCAM; Casali et al. 2007) and aphotometric system described in Hewett et al. (2006). The c (cid:13) , 000–000 Ben Burningham et al.
Figure 16.
Spectra for ULAS0034, CFBDS0059, ULAS1238 and ULAS1335 compared with the near-infrared spectrum of Jupiter(J. T.Rayner, M. C. Cushing, & W. D. Vacca 2008, in preparation) and laboratory transmission spectra of methane (Cruikshank & Binder1969) and ammonia (Irwin et al. 1999). The vertical green dashed lines highlight narrow features that appear to be common to ULAS1238,ULAS1335 and Jupiter. The vertical red dashed line highlights an interesting feature in the Jovian spectrum that is absent from theT dwarf spectra (see text). The vertical blue dotted line indicates the location of the red edge of the broad absorption attributed toammonia by Delorme et al. (2008).Object SpType T eff log g mass age distanceK log(cm s − ) M J Gyr pcULAS1017 T8p 750–850 5.0–5.5 33–70 1.6–15 31–54ULAS1335 T9 550–600 < Table 8.
A summary of our estimates for the properties of ULAS1017 and ULAS1335. Estimates for ULAS1238 have been generallyneglected in the absence of full flux calibrated spectral coverage. Distances assume that the dwarfs are single objects. pipeline processing and science archive are described in Ir-win et al (2008) and Hambly et al. (2008). We have useddata from the 3rd data release, which is described in de-tail in Warren et al. (in prep). Based on observations madewith the European Southern Observatory telescopes ob-tained from the ESO/ST-ECF Science Archive Facility. Re-sults reported here are based on observations obtained at theGemini Observatory under program numbers GN-2007B-Q- 26 and GN-2008A-Q-15. Gemini Observatory is operatedby the Association of Universities for Research in Astron-omy, Inc. (AURA), under a cooperative agreement with theNSF on behalf of the Gemini partnership: the National Sci-ence Foundation (United States), the Science and Technol-ogy Facilities Council (United Kingdom), the National Re-search Council (Canada), CONICYT (Chile), the AustralianResearch Council (Australia), Ministrio da Cincia e Tec- c (cid:13) , 000–000 ery cool T dwarfs REFERENCES
Allard F., Homeier D., Freytag B., Guillot T., Chabrier G.,Ludwig H., Allard N., Ferguson J., Barman T., HauschildtP., 2008, A&A, in prep.Burgasser A. J., Geballe T. R., Leggett S. K., KirkpatrickJ. D., Golimowski D. A., 2006, ApJ, 637, 1067Burgasser A. J., Kirkpatrick J. D., Brown M. E., Reid I. N.,Burrows A., Liebert J., Matthews K., Gizis J. E., DahnC. C., Monet D. G., Cutri R. M., Skrutskie M. F., 2002,ApJ, 564, 421Burgasser A. J., Liu M. C., Ireland M. J., Cruz K. L.,Dupuy T. J., 2008, ArXiv e-prints, 803Burrows A., Burgasser A. J., Kirkpatrick J. D., Liebert J.,Milsom J. A., Sudarsky D., Hubeny I., 2002, ApJ, 573,394Burrows A., Hubbard W. B., Lunine J. I., Liebert J., 2001,Reviews of Modern Physics, 73, 719Casali M., Adamson A., Alves de Oliveira C., Almaini O.,Burch K., Chuter T., Elliot J., Folger M., Foucaud S.,Hambly N., Hastie M., Henry D., Hirst P., Irwin M., IvesD., Lawrence A., Laidlaw K., Lee D., Lewis J., Lunney D.,McLay S., Montgomery D., Pickup A., Read M., Rees N.,Robson I., Sekiguchi K., Vick A., Warren S., WoodwardB., 2007, A&A, 467, 777Cruikshank D. P., Binder A. B., 1969, Commun. LunarPlanet. Lab., Vol. 6, (No. 103), p. 275 - 288, 6, 275Delorme P., Delfosse X., Albert L., Artigau E., ForveilleT., Reyl´e C., Allard F., Homeier D., Robin A. C., WillottC. J., Liu M. C., Dupuy T. J., 2008, A&A, 482, 961Duquennoy A., Mayor M., 1991, A&A, 248, 485Dye S., Warren S. J., Hambly N. C., Cross N. J. G.,Hodgkin S. T., Irwin M. J., Lawrence A., Adamson A. J.,Almaini O., Edge A. C., Hirst P., Jameson R. F., LucasP. W., van Breukelen C., and 31 co-authors, 2006, MN-RAS, 372, 1227Fazio G. G., Hora J. L., Allen L. E., Ashby M. L. N.,Barmby P., Deutsch L. K., Huang J.-S., Kleiner S.,Marengo M., Megeath S. T., 55 co-authors, 2004, ApJS,154, 10Golimowski D. A., Leggett S. K., Marley M. S., Fan X.,Geballe T. R., Knapp G. R., Vrba F. J., Henden A. A.,Luginbuhl C. B., Guetter H. H., Munn J. A., Canzian B.,Zheng W., and six co-authors, 2004, AJ, 127, 3516Hambly N. C., Collins R. S., Cross N. J. G., Mann R. G.,Read M. A., Sutorius E. T. W., Bond I., Bryant J., Emer-son J. P., Lawrence A., Rimoldini L., Stewart J. M., Williams P. M., Adamson A., Hirst P., Dye S., WarrenS. J., 2008, MNRAS, 384, 637Hewett P. C., Warren S. J., Leggett S. K., Hodgkin S. T.,2006, MNRAS, 367, 454Hodapp K. W., Jensen J. B., Irwin E. M., Yamada H.,Chung R., Fletcher K., Robertson L., Hora J. L., SimonsD. A., Mays W., Nolan R., Bec M., Merrill M., FowlerA. M., 2003, PASP, 115, 1388Irwin P. G. J., Calcutt S. B., Sihra K., Taylor F. W., WeirA. L., Ballard J., Johnston W. B., 1999, Journal of Quan-titative Spectroscopy and Radiative Transfer, 62, 193Kendall T. R., Tamura M., Tinney C. G., Mart´ın E. L.,Ishii M., Pinfield D. J., Lucas P. W., Jones H. R. A.,Leggett S. K., Dye S., Hewett P. C., Allard F., BaraffeI., Barrado Y Navascu´es D., Carraro G., Casewell S. L.,Chabrier G., Chappelle R. J., Clarke F., Day-Jones A.,Deacon N., Dobbie P. D., Folkes S., Hambly N. C.,Hodgkin S. T., Nakajima T., Jameson R. F., LodieuN., Magazz`u A., McCaughrean M. J., Pavlenko Y. V.,Tadashi N., Zapatero Osorio M. R., 2007, A&A, 466, 1059Kirkpatrick J. D., Reid I. N., Liebert J., Cutri R. M., Nel-son B., Beichman C. A., Dahn C. C., Monet D. G., GizisJ. E., Skrutskie M. F., 1999, ApJ, 519, 802Kobayashi N., Tokunaga A. T., Terada H., Goto M., We-ber M., Potter R., Onaka P. M., Ching G. K., YoungT. T., Fletcher K., Neil D., Robertson L., Cook D., Iman-ishi M., Warren D. W., 2000, in Presented at the Societyof Photo-Optical Instrumentation Engineers (SPIE) Con-ference, Vol. 4008, Proc. SPIE Vol. 4008, p. 1056-1066,Optical and IR Telescope Instrumentation and Detectors,Masanori Iye; Alan F. Moorwood; Eds., Iye M., MoorwoodA. F., eds., pp. 1056–1066Lawrence A., Warren S. J., Almaini O., Edge A. C., Ham-bly N. C., Jameson R. F., Lucas P., Casali M., AdamsonA., and thirteen co-authors, 2007, MNRAS, 379, 1599Leggett S. K., Currie M. J., Varricatt W. P., HawardenT. G., Adamson A. J., Buckle J., Carroll T., Davies J. K.,Davis C. J., Kerr T. H., Kuhn O. P., Seigar M. S., WoldT., 2006, MNRAS, 373, 781Leggett S. K., Golimowski D. A., Fan X., Geballe T. R.,Knapp G. R., Brinkmann J., Csabai I., Gunn J. E., Haw-ley S. L., Henry T. J., Hindsley R., Ivezi´c ˇZ., Lupton R. H.,Pier J. R., Schneider D. P., Smith J. A., Strauss M. A.,Uomoto A., York D. G., 2002, ApJ, 564, 452Leggett S. K., Marley M. S., Freedman R., Saumon D., LiuM. C., Geballe T. R., Golimowski D. A., Stephens D. C.,2007a, ApJ, 667, 537Leggett S. K., Saumon D., Marley M. S., Geballe T. R.,Golimowski D. A., Stephens D., Fan X., 2007b, ApJ, 655,1079Leinert C., Zinnecker H., Weitzel N., Christou J., RidgwayS. T., Jameson R., Haas M., Lenzen R., 1993, A&A, 278,129Lodieu N., Pinfield D. J., Leggett S. K., Jameson R. F.,Mortlock D. J., Warren S. J., Burningham B., LucasP. W., Chiu K., Liu M. C., Venemans B. P., McMahonR. G., Allard F., Baraffe I., Y Navascu´es D. B., CarraroG., Casewell S. L., Chabrier G., Chappelle R. J., ClarkeF., Day-Jones A. C., Deacon N. R., Dobbie P. D., FolkesS. L., Hambly N. C., Hewett P. C., Hodgkin S. T., JonesH. R. A., Kendall T. R., Magazz`u A., Mart´ın E. L., Mc-Caughrean M. J., Nakajima T., Pavlenko Y., Tamura M., c (cid:13) , 000–000 Ben Burningham et al.
Tinney C. G., Zapatero Osorio M. R., 2007, MNRAS, 379,1423Looper D. L., Gelino C. R., Burgasser A. J., KirkpatrickJ. D., 2008, ArXiv e-prints, 803Looper D. L., Kirkpatrick J. D., Burgasser A. J., 2007, AJ,134, 1162Manchado A., Fuentes F. J., Prada F., Ballesteros E., Bar-reto M., Carranza J. M., Escudero I., Fragoso-Lopez A. B.,Joven-Alvarez E., Manescau A., Pi M., Rodriguez-RamosL. F., Sosa N. A., 1998, in Presented at the Society ofPhoto-Optical Instrumentation Engineers (SPIE) Confer-ence, Vol. 3354, Proc. SPIE Vol. 3354, p. 448-455, In-frared Astronomical Instrumentation, Albert M. Fowler;Ed., Fowler A. M., ed., pp. 448–455Marley M. S., Saumon D., 2008, Private CummunicationMarley M. S., Seager S., Saumon D., Lodders K., AckermanA. S., Freedman R. S., Fan X., 2002, ApJ, 568, 335Moorwood A., Cuby J.-G., Biereichel P., Brynnel J., De-labre B., Devillard N., van Dijsseldonk A., Finger G., 13co-authors, 1998, The Messenger, 94, 7Patten B. M., Stauffer J. R., Burrows A., Marengo M.,Hora J. L., Luhman K. L., Sonnett S. M., Henry T. J.,Raghavan D., Megeath S. T., Liebert J., Fazio G. G., 2006,ApJ, 651, 502Pinfield D. J., Burningham B., Tamura M., Leggett S. K.,Lodieu N., Lucas P. W., Mortlock D. J., Warren S. J.,Homeier D., Ishii M., McMahon R. G., and 29 co-authors,2008, MNRAS submittedRoche P. F., Lucas P. W., Mackay C. D., Ettedgui-AtadE., Hastings P. R., Bridger A., Rees N. P., Leggett S. K.,Davis C., Holmes A. R., Handford T., 2003, in Pre-sented at the Society of Photo-Optical InstrumentationEngineers (SPIE) Conference, Vol. 4841, Instrument De-sign and Performance for Optical/Infrared Ground-basedTelescopes. Edited by Iye, Masanori; Moorwood, Alan F.M. Proceedings of the SPIE, Volume 4841, pp. 901-912(2003)., Iye M., Moorwood A. F. M., eds., pp. 901–912Saumon D., Marley M. S., Cushing M. C., Leggett S. K.,Roellig T. L., Lodders K., Freedman R. S., 2006, ApJ,647, 552Saumon 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, 1136Vrba F. J., Henden A. A., Luginbuhl C. B., Guetter H. H.,Munn J. A., Canzian B., Burgasser A. J., KirkpatrickJ. D., Fan X., Geballe T. R., Golimowski D. A., KnappG. R., Leggett S. K., Schneider D. P., Brinkmann J., 2004,AJ, 127, 2948Warren S. J., Mortlock D. J., Leggett S. K., Pinfield D. J.,Homeier D., Dye S., Jameson R. F., Lodieu N., LucasP. W., Adamson A. J., and 14 co-authors, 2007, MNRAS,381, 1400 c (cid:13)000