한국   대만   중국   일본 
WD 0346+246 in Taurus
The Wayback Machine - https://web.archive.org/web/20060828155306/http://ecf.hq.eso.org:80/~ralbrech/novdec97apjl/975638.html
T HE A STROPHYSICAL J OURNAL , 489:L157 –L160, 1997 November 10
© 1997. The American Astronomical Society. All rights reserved. Printed in U.S.A.
Up: Issue Table of Contents
Go to: Search Page | Previous Article | Next Article
Other formats: HTML (small files) | PDF (335 kb) | PostScript pages (951 kb)

WD 0346+246: A Very Low Luminosity, Cool Degenerate in Taurus

N. C. H AMBLY

Royal Observatory, Blackford Hill, Edinburgh, EH9 3HJ, Scotland, UK; N.Hambly@roe.ac.uk

S. J. S MARTT

Isaac Newton Group of Telescopes, Apartado de Correos 321, 38780 Santa Cruz de La Palma, La Palma, Islas Canarias, Spain; sjst@ing.iac.es

AND

S. T. H ODGKIN

Astronomy Group, Department of Physics and Astronomy, Leicester University, University Road, Leicester, LE1 7RH, England, UK; sth@star.le.ac.uk">

sth@star.le.ac.uk

Received 1997 August 18; accepted 1997 September 11; published 1997 October 13


ABSTRACT

     We report the serendipitous discovery of a new, very low luminosity, cool degenerate in the region of Taurus. The object was found as a very high proper-motion star ( μ = 1 &farcs;3 yr -1 ) on seven I -band UK Schmidt Telescope plates, dating from 1987 to 1994, via digitized scans from the new, fast, high-precision microdensitometer SuperCOSMOS. Photometry and spectrophotometry indicate that the object has a temperature comparable to those of the handful of coolest white dwarfs currently known ( T ∼3900 K). We discuss the relevance of this discovery to current research concerning Galactic structure and evolution.

Subject headings: Galaxy: stellar content —stars: individual (WD 0346+246) —white dwarfs

CONTENTS

§1.  INTRODUCTION

     White dwarfs (WDs), the evolutionary end state of all intermediate- and low-mass stars, have been the subject of much discussion in the modern astronomical literature (e.g., Isern, Hernanz, & García-Berro 1997 , and references therein). Their relative structural simplicity (at least for the hotter degenerates) makes them attractive targets for theoretical astrophysics —for instance, the famous derivation of the maximum WD mass ( Chandrasekhar 1939 ). Old and cool WDs can tell us much about the physics of degenerate matter and the evolutionary history of the Galaxy. Recently, WDs have become relevant in two major aspects of Galactic research:

     1. A lower limit to the age of the Galactic disk is constrained by the age of the oldest (and therefore coolest) WDs present. Since the pioneering work of Weidemann (1967) , there have been several increasingly sophisticated attempts to measure the luminosity function (LF) for disk WDs in order to estimate this age via models of their cooling (e.g., Liebert, Dahn, & Monet 1988 , and references therein; Oswalt et al. 1996 ; Knox & Hawkins 1997 ; Bergeron, Ruiz, & Leggett 1997 , hereafter BRL97 ). While it is relatively simple to identify hot WDs in the local Galactic disk, obtaining a complete and unbiased sample of cool degenerates is fraught with difficulty because of the vast numbers of contaminating G, K, and M dwarfs. The usual method is to use reduced proper-motion diagrams (e.g., Evans 1992 , and references therein) to identify faint objects with relatively high proper motion. Only with spectroscopic confirmation can the objects so selected be classified as cool degenerates. With the exception of Knox & Hawkins (1997) , the above LF determinations have, to a large extent, relied on existing nearby star/proper-motion catalogs, especially those of Luyten (e.g., Luyten 1979, and references therein), along with estimates of completeness. The completeness of Luyten's monumental surveys has always been difficult to assess (to quote Luyten himself, “…the question of completeness is a vexing one …”). Consequently, there has been some debate as to the origin of the turnover in the WD LF: the absence of WDs having M $\mathstrut{_{V}}$>16 is possibly a selection effect, despite rigorous attempts to quantify the completeness (e.g., Dawson 1986 ). Indeed, both Ruiz (1996) and Oswalt et al. (1996) find an increase of a factor ∼5 over Liebert et al. (1988) in the space density of the coolest WDs.

     2. Recent results from the MACHO microlensing project ( Alcock et al. 1997 ) show an increasing number of long timescale events that could indicate the presence of higher mass lensing objects when compared with previous analyses of earlier results (e.g., Jetzer 1994 ; Han & Gould 1996 ). Despite problems of metal enrichment in the early stages of Galactic evolution from the progenitor population, the idea that WDs could cause a significant number of microlensing events has been discussed at some length (e.g., Gibson & Mould 1997 , and references therein)

     Here we report the discovery of a very low luminosity, cool degenerate. The object was identified via its large proper motion during work to find brown dwarfs in the Pleiades. Section 2 describes the observational properties, while § 3 discusses the significance of this discovery and § 4 summarizes our conclusions.

§2.  OBSERVATIONS

§2.1.  Astrometry

      Table 1 gives details of the Schmidt plates that were obtained as part of an ongoing project to detect brown dwarfs in the Pleiades cluster (e.g., Hambly, Hawkins, & Jameson 1991 ; Stauffer et al. 1997 ). The plates had hypersensitized Kodak IV-N emulsions and were exposed through a Schott RG 715 filter for 90 minutes on the UK Schmidt Telescope at Siding Springs Observatory, New South Wales, Australia. The field center of all plates was 3 h 44 m + 23°57 &arcmin; (B1950.0). All plates were scanned using the new, fast, high-precision microdensitometer SuperCOSMOS at the Royal Observatory, Edinburgh (e.g., Hambly et al. 1997 ). This machine digitizes the 6 $\mathstrut{^{{\circ}}}$×6 $\mathstrut{^{{\circ}}}$ area of the Schmidt plate in ∼2 hr, with 10 μm (0 &farcs;67) pixels at 15 bit gray-level resolution, producing ∼2 Gbyte of data. The digitized map is then thresholded and examined for connected pixels to produce a parameterized image catalog. Rethresholding at eight higher levels deblends multiple images ( Beard, MacGillivray, & Thanisch 1990 ). The data presented here were thresholded at 3 σ above sky, and a minimum area cut of five connected pixels was used, resulting in typically ∼300,000 images detections on each plate. Image pairing was achieved by choosing a master plate (in this case plate I16468) and transforming the coordinate system of the others onto the master frame. Using a multiple pass procedure, an increasing pairing radius was employed to pair images on the slave plates to those on the master. The criterion for an image to be paired was simply that the nearest image on any slave plate to that on the master was considered paired, provided that the shape and magnitude parameters were consistent within the known plate-to-plate differences between master and slave. Relative shifts for all images appearing on all seven plates were then determined by fitting local linear least-squares plate models in 16×16 subplate areas for all images with respect to the mean positions from all seven plates. Weighted linear least-squares fits to these shifts as a function of time then yielded the relative proper motion for each image. The standard errors for each data point were simply the rms residuals, as a function of magnitude, from the local linear least-squares model. For a typical sky-limited, fine-grained Schmidt plate measured on SuperCOSMOS, these errors range from 0.3 μm for bright images to more than 3 μm for images at the detection limit (e.g. , Hambly et al. 1997 ).

     One object (hereafter WD 0346+246) stood out as an extremely high proper-motion faint star that does not appear in the Luyten catalogs. Figure 1 shows the object's change in position with time. The astrometric parameters of the object are presented in Table 2 . Figure 2 presents images of a 4×4 arcminute region around the epoch 1994.99 position of WD 0346+246. Figure 2 a shows a red passband image at epoch 1951.91 from the POSS I plate E441. Figure 2 b is an I passband image at epoch 1987.91, while Figure 2 c is at epoch 1994.99. The high proper motion is easily discernible in these images.

Fig. 1 Fig. 2

§2.2.  Photometry

     Photometry of WD 0346+246 was obtained on the night of 1997 February 16 using the 1.0 m Jacobus Kapteyn Telescope on the island of La Palma. The standard BVRI filter set was used, and observations of standard stars from the lists of Landolt (1992) were made. The weather was stable, but with high extinction as the result of dust. The data were reduced within the IRAF environment following standard procedures (debiasing, flat fielding, and aperture photometry). The exposure time for the B frame was sufficient only to put an upper limit on the object's B magnitude. The colors are listed in Table 2 .

§2.3.  Spectroscopy

     Spectroscopy of WD 0346+246 was obtained on the night of 1997 February 7 using the intermediate dispersion spectroscopic and imaging system (ISIS) on the 4.2 m William Herschel Telescope on the island of La Palma. Using gratings R158R and R158B on the red and blue channels respectively gave a spectral coverage of 3500 –8500 Å, with 2.9 Å per CCD pixel and resolution ∼10 Å. A 30 minute exposure was made using a 1 &arcsec; slit; wide-slit observations of the target and the spectrophotometric standard Hz 14 ( Massey et al. 1988 ) were also made to provide the flux calibration. The bright star HR 1864 was also observed to provide telluric absorption corrections redward of 7000 Å. The data were again reduced within the IRAF environment, following standard procedures. A nominal extinction correction was applied to the data since only one standard was observed and no independent extinction derivation was possible. The spectrophotometry is presented in Figure 3 . Also shown ( dashed lines ) are blackbody spectra at T =3900±200 K, normalized to the stellar continuum flux at 6500 Å, to illustrate (albeit very approximately) the effective temperature of the star.

Fig. 3

§3.  DISCUSSION

     In Figure 4 we plot a reduced proper-motion diagram for the comprehensive sample of known cool degenerates analyzed by BRL97 . The coolest and least luminous objects are identified; clearly, the object is among the few coolest WDs known. The question of whether WD 0346+246 has a record-breaking low temperature and luminosity must await further observations. For example, in the discovery paper for ER 8 ( Ruiz et al. 1986 ), an estimated distance of ∼5 pc and a blackbody temperature of T ∼3500 K were found; however, the distance has subsequently been determined as d =15.0±0.5 pc ( Ruiz et al. 1990 ), while a model atmosphere fitted to accurate broadband BVRIJHK photometry indicates T =4170±70 K ( BRL97 ). Also, ESO 439-26 ( Ruiz, Anguita, & Maza 1989 ) was originally thought to be extremely cool as a result of its intrinsic luminosity as implied by the parallax; subsequently it was shown to have a very low luminosity because of its high mass and consequent small radius, with a model atmosphere analysis of broadband colors yielding T >4200 K. Use of the model atmosphere computations of Bergeron, Wesemael, & Beauchamp (1995) in conjunction with the photometry and an assumed gravity of $\mathstrut{{\rm log}}$g =8 allows exploration of other physical parameters of this star. For example, assuming a pure helium atmosphere, a match is found for mass m ∼0.57 M $\mathstrut{_{{\odot}}}$ , T $\mathstrut{_{{\rm eff}}}$=4500 K, and age ∼7.8 Gyr; assuming a pure hydrogen atmosphere (as Bergeron et al. point out, the absence of Balmer lines in the spectrum of such a cool object does not preclude the possibility of a hydrogen atmosphere), we find a match for m ∼0.58 M $\mathstrut{_{{\odot}}}$, T $\mathstrut{_{{\rm eff}}}$=4000 K, and age ∼8.5 Gyr. In either case, the implied distance is ∼40 pc, yielding a tangential velocity of v $\mathstrut{_{t}}$∼250 km s -1 , indicative of the Galactic halo population. On the other hand, assuming old-disk kinematics ( v $\mathstrut{_{t}}$=70 km s -1 ) requires the distance to be only ∼10 pc for such a large proper motion: WD 0346+246 would then be a remarkably low luminosity object. Obviously, without an intrinsic luminosity estimate (via a parallax measurement) to constrain the absolute magnitude, these comparisons do little more than illustrate the possibilities.

Fig. 4

      BRL97 have shown that it is possible to see chemical composition and mass effects in the evolution of cool WDs given good quality data (i.e., accurate photometry and parallaxes) via model atmosphere analysis. Clearly, it is difficult to estimate the absolute luminosity of WD 0346+246 given such effects and the intrinsic spread in space velocity in the cool degenerate sample presented in Figure 4 . However, the object is nearly 1 mag fainter than ER 8 in the reduced proper-motion diagram; perhaps it indeed has a higher mass (and therefore smaller radius) —cf. ESO 439-26 ( Ruiz et al. 1995 ). Any systematic errors leading to an underestimate in the object's temperature would reinforce such an interpretation. Alternatively, as already stated, the object may be a halo WD with a correspondingly large space velocity, as perhaps is LHS 282 (unfortunately, the absence of any spectral lines precludes a radial velocity measurement in such cool degenerates). It is interesting to speculate that perhaps it is objects such as these that are causing the long timescale microlensing events that have recently been reported ( Alcock et al. 1997 ).

     Finally, we note that such cool degenerates as WD 0346+246 are likely to be reasonably numerous, and it is faint, high proper-motion incompleteness in nearby star catalogs such as the LHS and NLTT that results in so few being currently known. For example, assuming this object is similar to ER 8, then its distance is likely to be ∼36 pc and its bolometric luminosity $\mathstrut{\mathop{{\rm log}}_{10}\nolimits }$&parl0;L &solm0;L $\mathstrut{_{{\odot}}}$&parr0;∼-4.4. In the bolometric luminosity range -4.25> $\mathstrut{\mathop{{\rm log}}_{10}\nolimits }$&parl0;L &solm0;L $\mathstrut{_{{\odot}}}$&parr0;>-4.5 the WD LF of Oswalt et al. (1996) predicts a total space density of $\mathstrut{\mathop{{\rm log}}_{10}\nolimits }$Φ=-2.8×10 $\mathstrut{^{-3}}$ pc -3 . In a spherical volume of 40 pc, this corresponds to 424 stars, or ∼10 -2 per square degree. Hence, if a handful of Schmidt fields are searched down to m ∼20, one of these μ>1 &arcsec; yr -1 cool degenerates should turn up. It is worth stating that a search for high proper-motion objects using the first and second epoch Schmidt telescope sky surveys (e.g., Morgan et al. 1992 ) is by no means a trivial task. In the northern hemisphere in particular, the epoch interval is ∼40 yr; consequently, any object having μ>1 &arcsec; yr -1 will have moved by up to 1 &arcmin;, and great care will be needed when pairing images. The inevitable spurious images and spurious image pairings that occur when working with just two plates will no doubt compound these problems.

§4.  CONCLUSIONS

     We have discovered a high proper-motion object that follow-up photometry and spectroscopy have shown to be an extremely cool degenerate and possibly the coolest degenerate currently known. Accurate position and proper-motion data have been presented that will enable this object to be included in the ongoing cool degenerate investigations such as that of BRL97 . A parallax measurement along with a model atmosphere fitted to accurate UBVRIJHK photometry is now needed in order to clarify whether this star is indeed the coolest and lowest luminosity WD currently known.

ACKNOWLEDGMENTS

     It is a pleasure to acknowledge once again the work of the UK Schmidt Telescope Unit in obtaining the high-quality plate material used here. Also, we wish to thank Harvey MacGillivray and Eve Thomson for scanning the plates on SuperCOSMOS. N. C. H. acknowledges useful discussions with Richard Knox, Mike Hawkins, René Rutten, Lance Miller, and John Peacock. S. T. H. acknowledges support from the UK PPARC, and we thank Martin Cossburn and Nic Walton for help with the JKT photometry and WHT spectroscopy, respectively. Data reduction was undertaken on the Leicester University and Royal Observatory Edinburgh Starlink Nodes. The William Herschel and Jacobus Kapteyn Telescopes on the island of La Palma are operated by the Royal Greenwich Observatory, on behalf of the UK PPARC, and at the Spanish Observatorio del Roque de los Muchachos of the Instituto de Astrofisica de Canarias. The Palomar Observatory first epoch Sky Survey was funded by a grant from the National Geographic Society to the California Institue of Technology. Finally, we thank the referee for a prompt and useful report.

REFERENCES

FIGURES


Full image (53kb) | Discussion in text

     F IG . 1. —Relative proper-motion determination for WD 0346+246.


Full image (51kb) | Discussion in text

     F IG . 2. —Finders for WD 0346+246 at ( a ) epoch 1951.91, ( b ) 1987.91, and ( c ) 1994.99. ( a ) Red passband image; ( b ) and ( c ) I passband images.


Full image (54kb) | Discussion in text

     F IG . 3. —Spectrum of WD 0346+246. Dashed lines are blackbody spectra at T =3900±200 K, normalized to the observed flux at 6500 Å.


Full image (47kb) | Discussion in text

     F IG . 4. —Reduced proper-motion diagram for the BRL97 sample of cool WDs. The coolest and least luminous objects are labeled. The filled triangle with error bars represents WD 0346+246.

TABLES

TABLE 1
P LATE M ATERIAL
Plate Number LST Epoch
I12259... 03:08 1987.910
I13497... 03:36 1989.962
I15278... 03:08 1992.997
I16395... 02:40 1994.858
I16448... 03:20 1994.934
I16464... 02:54 1994.986
I16468... 02:59 1994.989

Image of typeset table | Discussion in text
TABLE 2
O BSERVATIONAL P ARAMETERS
R.A. a ... 3 h 46 m 46 &fs;30
Decl. a ... + 24°56 &arcmin;08 &farcs;6
I ... 17.68 ± 0.06
( R - I )... 0.65±0.08
V ... 19.04±0.12
B ... > 20.3
μ(arcsec yr -1 )... 1.27±0.04
P.A. (E of N)... 155 $\mathstrut{^{{\circ}}}$
T $\mathstrut{_{{\rm eff}}}$ (K) (blackbody)... ∼3900±200

      a Equinox J2000.0, epoch 1994.99, ± 0 &farcs;3.

Image of typeset table | Discussion in text
Up: Issue Table of Contents
Go to: Top of This Article | Search Page | Previous Article | Next Article