Draft version August 10, 2015 Preprint typeset using LATEX style emulateapj v. 11/10/09 FUNDAMENTAL PARAMETERS AND SPECTRAL ENERGY DISTRIBUTIONS OF YOUNG AND FIELD AGE OBJECTS WITH MASSES SPANNING THE STELLAR TO PLANETARY REGIME Joseph C. Filippazzo1,2,3, Emily L. Rice1,2,3, Jacqueline Faherty2,5,6, Kelle L. Cruz2,3,4, Mollie M. Van Gordon7, Dagny L. Looper8 1Department of Engineering Science and Physics, College of Staten Island, City University of New York, 2800 Victory Blvd, Staten Island, NY 10314, USA 2Department of Astrophysics, American Museum of Natural History, New York, NY 10024, USA 3The Graduate Center, City University of New York, New York, NY 10016, USA 4Department of Physics and Astronomy, Hunter College, City University of New York, New York, NY 10065, USA 5Department of Terrestrial Magnetism, Carnegie Institution of Washington, DC 20015, USA 6Hubble Fellow 7Department of Geography, University of California, Berkeley, CA 94720, USA and 8Tisch School of the Arts, New York University, New York, NY 10003, USA Draft version August 10, 2015 ABSTRACT We combine optical, near-infrared and mid-infrared spectra and photometry to construct expanded spectral energy distributions (SEDs) for 145 field age (>500 Myr) and 53 young (lower age estimate <500 Myr) ultracool dwarfs (M6-T9). This range of spectral types includes very low mass stars, brown dwarfs, and planetary mass objects, providing fundamental parameters across both the hydrogen and deuterium burning minimum masses for the largest sample assembled to date. A subsample of 29 objects have well constrained ages as probable members of a nearby young moving group (NYMG). We use 182 parallaxes and 16 kinematic distances to determine precise bolometric luminosities (Lbol) and radius estimates from evolutionary models give semi-empirical effective temperatures (Teff) for the full range of young and field age late-M, L and T dwarfs. We construct age-sensitive relationships of luminosity, temperature and absolute magnitude as functions of spectral type and absolute magnitude to disentangle the effects of degenerate physical parameters such as Teff, surface gravity, and clouds on spectral morphology. We report bolometric corrections in J for both field age and young objects and find differences of up to a magnitude for late-L dwarfs. Our correction in Ks shows a larger dispersion but not necessarily a different relationship for young and field age sequences. We also characterize the NIR-MIR reddening of low gravity L dwarfs and identify a systematically cooler Teff of up to 300K from field age objects of the same spectral type and 400K cooler from field age objects of the same MH magnitude. Subject headings: brown dwarfs, stars: low-mass, stars: fundamental parameters 1. INTRODUCTION Burrows et al. 2011; Barman 2008) account for more Brown dwarfs are unable to sustain nuclear fusion in complex chemistry and dynamics than ever before, in- their cores due to insufficient mass and are thus degen- complete physics and line lists frequently limit reliable erate across effective temperature, mass, and age. While data fits to those brown dwarfs which exhibit the simplest these objects all contract to about the size of Jupiter atmospheric conditions. Even for these objects, there within 500 Myr (Baraffe et al. 1998), the extended pho- are broad regions of the model spectra that have yet to tospheres of younger objects introduce the radius as yet reproduce observations and so must be excluded from another elusive observable. Entanglement of these fun- the fitting routine (e.g. Cushing et al. 2008; Mann et al. 2015). Consequently, derivation of fundamental param- arXiv:1508.01767v1 [astro-ph.SR] 7 Aug 2015 damental parameters prohibits precise atmospheric char- acterization by spectral type and color alone. This neces- eters with model atmospheres depend heavily on the in- sitates determination of broader physical quantities such cluded wavelength ranges, the resolution of the spectrum, as distance, radius, and luminosity. Flux calibrated spec- the fitting technique, and the models used. tral energy distributions (SEDs) comprised of spectra as Rice et al. (2010a) fit model atmospheres to young well as photometry enable precise empirical determina- (<10Myr) and field age late-M dwarfs and concluded that a combination of medium- and high-resolution NIR tion of Lbol which can then be used to estimate additional stellar parameters. spectra is needed to derive fundamental parameters, Effective temperatures lower than about 3000K cause though this is rarely attempted due to limited avail- the emergent spectra of brown dwarfs to deviate sub- able data especially for fainter sources. Patience et al. stantially from that of a blackbody due to absorption (2012) fit five model atmosphere grids to NIR spectra and scattering from molecules, dust and clouds. Deter- of a small sample of 1-50 Myr M8-L5 dwarfs and found mination of their physical properties is further compli- Teff discrepant by up to 300K based on the models used. cated as these substellar objects age and cool, chang- Despite such documented inconsistencies, this remains ing opacity sources and evolving through later spec- the most common way to extract atmospheric properties tral types. Though current ultracool dwarf model at- such as effective temperature, surface gravity, sedimen- mosphere codes (Allard 2013; Saumon & Marley 2008; tation efficiency, and metallicity (Cushing et al. 2008; 2 Filippazzo et al. Stephens et al. 2009; Witte et al. 2011; Bonnefoy et al. properties of ultracool dwarfs. Conclusions are presented 2014; Dieterich et al. 2014; Manjavacas et al. 2014). Re- in Section 10. cent Lbol determinations of large samples of M, L and 2. THE SAMPLE T dwarfs (Stephens et al. 2009; Dupuy & Kraus 2013; Dieterich et al. 2014; Schmidt et al. 2014) have used Our goal was to assemble nearly complete SEDs to in- MIR spectroscopy or photometry but have similarly em- vestigate global trends in the fundamental parameters ployed various model atmospheres and fitting routines to of substellar objects, which presumably form like stars estimate flux in areas without spectral coverage. Most from the collapse of a shocked cloud of cold gas and other Lbol estimates in the literature use bolometric cor- dust. To accomplish this we constructed a sample of rections derived from these samples. These techniques diskless objects no longer associated with a molecular are self-consistent in the parameters they predict but suf- cloud with masses just above the hydrogen burning min- fer from both known and unidentified systematics intro- imum mass of about 79MJup all the way down to just duced by imperfect model atmosphere codes. Until the below the deuterium burning minimum mass of about model grids reproduce the variety of our observations and 12MJup (Burrows et al. 1997; Chabrier et al. 2000). All fitting routines become more robust, a strictly empirical objects were required to be within 80pc to avoid inter- sanity check is needed. stellar extinction effects (Aumer & Binney 2009). Direct integration of flux calibrated SEDs provides Young objects are distinguished by their very red − Lbol as a function of more ubiquitous measurements such J KS color due to dust and clouds in or above as magnitude, color, spectral index, and spectral type. the photosphere (Kirkpatrick et al. 2006; Looper et al. Accurate characterization of these distance-scaled rela- 2008b), though not all very red objects have youth in- tionships can be powerful tools for inferring the atmo- dicators (Faherty et al. 2013). Therefore young dwarfs spheric properties of additional ultracool dwarfs. Since were identified based on probable membership in a distance is the dominant source of uncertainty in these NYMG (Faherty et al. in prep; Reidel et al. in prep; calculations, the accumulation of trigonometric paral- Gagn´eet al. 2015a). Additional young L0-L5 dwarfs laxes for a diverse sample of late-M, L and T dwarfs (e.g. were identified as those with a β or γ spectral type suf- Dahn et al. 2002; Tinney et al. 2003; Vrba et al. 2004; fix (Kirkpatrick 2005; Kirkpatrick et al. 2006; Cruz et al. Faherty et al. 2009; Dupuy & Liu 2012; Marocco et al. 2009; Rice et al. 2010b, Cruz et al. in prep) indicating 2013; Dieterich et al. 2014; Tinney et al. 2014) is crucial low surface gravity features in the optical such as weak to providing precise Lbol measurements across the en- alkali doublets and weak metal hydride absorption. Late- tire stellar/brown dwarf/planetary mass sequence. Ra- M and late-L dwarf spectral types have been updated dius and mass can then be inferred from from evolution- with β/γ suffixes to reflect intermediate and very low ary models and Teff can be calculated from the Stefan- surface gravity identified in the NIR (Allers & Liu 2013). Boltzmann Law. This is preferable to deriving Lbol from In total, the sample contains 29 NYMG members and 24 Teff values obtained by model atmosphere fits since the low gravity dwarfs that were not placed in a NYMG. results are not tied to the fidelity of a fitting routine, the We grouped all objects into three subsamples: 1) the complexities of modeled atmospheric conditions, or the core sample of 28 objects with a parallax measurement, quality of the data. optical through MIR photometry, and optical through Construction of ultracool dwarf SEDs from nearly com- MIR spectra, 2) the extended sample of 154 objects with plete observational coverage is therefore ideal and timely the base requirements of a parallax, NIR spectrum, NIR due to the mid-infrared (MIR) photometry of the Wide- photometry, and MIR photometry, and 3) the kinematic Field Infrared Survey Explorer (WISE; Wright et al. sample of 16 young objects with the same photomet- 2010) and the Spitzer Space Telescope Infrared Ar- ric and spectral requirements as the extended sample ray Camera (IRAC; Fazio et al. 2004), as well as the but with distances constrained by kinematics based on groundswell of publicly available optical and infrared NYMG membership (Faherty et al., in prep).