1: %
2: % file /scr/jlc/m31_hires/hires_sep2006/apjl_accept_oct2006/ms.tex
3: % edited version retrieved from the ApJ web site
4: %
5:
6: %\documentclass[11pt,preprint]{aastex}
7: \documentclass[12pt,preprint]{aastex}
8:
9:
10: \newcommand{\kms}{km~s$^{-1}$}
11: \newcommand{\etal}{{\it et al.\/}}
12: \newcommand{\subsun}{\mbox{$_{\odot}$}}
13:
14: \begin{document}
15:
16:
17: \title{The Not So Extraordinary Globular Cluster M31~037--B327\altaffilmark{1}}
18:
19:
20: \author{Judith G. Cohen\altaffilmark{2}}
21:
22: \altaffiltext{1}{Based on observations obtained at the
23: W.M. Keck Observatory, which is operated jointly by the California
24: Institute of Technology, the University of California, and the
25: National Aeronautics and Space Administration.}
26:
27: \altaffiltext{2}{Palomar Observatory, Mail Stop 105-24,
28: California Institute of Technology, Pasadena, Ca., 91125}
29:
30:
31: \begin{abstract}
32:
33: A velocity dispersion has been measured for the luminous
34: globular cluster M31~037--B327, claimed to be the
35: most massive star cluster in the Local Group and to
36: be a young ``super star cluster'' that has survived
37: to an old age. M31~037--B327
38: has a mass comparable to that of M31~G1, but not
39: significantly larger. Although near the upper end for
40: the mass distribution of globular clusters, it is not an
41: unprecedented extraordinary object.
42:
43:
44: \end{abstract}
45:
46: \keywords{galaxies: individual (M31), galaxies: star clusters}
47:
48: \section{Introduction}
49:
50: The globular cluster 037--B327 is an extremely red non-stellar
51: object close to the disk of M31. \cite{barmby02a} calculated
52: its reddening, using photometry in the compiled catalog
53: of M31 globular clusters of \cite{barmby00}, to be $E(B-V) = 1.30\pm0.04$~mag,
54: and from this inferred that it was extremely
55: luminous, with $M_V = -11.7$~mag, making it
56: the most luminous globular cluster in M31. The recent compilation
57: of data for M31 globular clusters by the Bologna group
58: \citep{galletti04} also confirms the unusual properties
59: of this object.
60:
61: Recently \cite{ma06a} have studied this cluster in more detail.
62: They compare their new multicolor photometry to theoretical
63: spectral energy distributions of varying ages to determine
64: the reddening and age of this cluster. They find
65: that it has a photometric mass of $3.0\pm0.5 \times 10^7 M$\subsun;
66: they then claim this object to be the most massive star cluster of any age
67: in the Local Group. This claim is based in part on data from
68: the compiled catalogs of \cite{barmby00} for the M31 system and of
69: \cite{harris96} for the Milky Way globular
70: cluster system. \cite{ma06a} predict that M31~037--B327
71: has a 1D velocity dispersion of $72\pm13$~\kms,~ far larger
72: than that of any other known M31 globular cluster. Velocity dispersions
73: for a number of the brightest globular clusters in M31
74: were determined by \cite{george}; the highest
75: value they measured was $25.1\pm 0.3$~\kms, for M31 G1.
76:
77:
78: \section{$\sigma_v$ for M31~037--B327}
79:
80: We have obtained high spectral resolution spectra with
81: HIRES at Keck \citep{vogt94} of M31~037--B327
82: and of M31~G1, as well as of several metal-rich giants
83: on the upper RGB of the Draco dwarf spheroidal galaxy
84: which served as template stars. We took two 1500 sec
85: exposures of the object of interest, one 600 sec exposure of the much
86: brighter (at $V$) object M31~G1, and several suitable spectra of
87: template red giants from the same run in early September 2006.
88: All these spectra were taken with a $1.1^{\arcsec} \times 7^{\arcsec}$
89: slit, giving a spectral resolution of 35,000 with 5 pixels/spectral
90: resolution element ($\sim$1.3~km sec$^{-1}$ pixel$^{-1}$).
91:
92: We chose for analysis
93: parts of three different echelle orders
94: with strong broadened spectral features visible in the
95: spectrum of M31~037--B327. Since the flux
96: for this object was dropping rapidly towards the blue,
97: but the number of strong lines was dropping rapidly
98: towards the red, this was something of a compromise.
99: Cutouts of the spectra in one of the regions analyzed,
100: that near H$\alpha$, are shown for
101: the two M31 globular clusters and one of the
102: template objects in Fig.~\ref{figure_vsigma}. It is immediately
103: apparent that $\sigma_v$ for M31~037--B327 is comparable
104: to that of M31~G1, but is not significantly larger.
105: This impression is sustained when one examines the strongest
106: lines in these spectra over their full useful wavelength range,
107: which for the object of interest is from about 5100 to 8350~\AA.
108:
109: We used the Fourier
110: quotient method of \cite{sargent} to determine $\sigma_v$.
111: We subsequently applied
112: an aperture correction factor of 1.14 following \cite{george}.
113: The resulting $\sigma_v$ are given in Table~\ref{table_vsigma}.
114: Our measured $\sigma_v$ for M31~G1 is in good agreement with that of
115: \cite{george}, $25.1\pm0.3$~\kms.
116:
117: \section{Comparison of M31~037--B327 With M31~G1}
118:
119: M31~G1, if one ignores M31~037--B327, is widely believed to
120: be the most luminous globular cluster
121: in M31, see, e.g. \cite{meylan01}. \cite{gebhardt} suggested,
122: based on HST/STIS spectroscopy, that it contains
123: a 20,000~$M$\subsun\ central black hole.
124: \cite{baumgardt} dispute this; they obtain a good fit to
125: all available data for G1 with their dynamical model, which does
126: not include an intermediate-mass central black hole.
127:
128:
129: \cite{ma06b} utilize an ACS image from HST to determine
130: the structural parameters of M31~037--B327, and discuss the possibility,
131: suggested earlier by \cite{barmby02a},
132: that this object is the nucleus of a dwarf galaxy accreted by M31.
133: The ACS image reveals a dust lane crossing the face
134: of the cluster. Toy models of a uniform extinction over
135: most of the cluster with much larger $E(B-V)$ over a
136: smaller part of the cluster suggest substantial errors
137: in the values of the dereddened fluxes
138: can occur in such circumstances if a standard
139: extinction curve as a function of wavelength is applied assuming
140: a constant $E(B-V)$.
141: The reddening of G1 is small as it has a projected distance from the center
142: of M31 of $\sim$40 kpc; we assume only the foreground
143: Galactic reddening applies.
144:
145: We compare the properties of M31~037--B327 with those of G1.
146: The values of $\sigma_v$ are comparable (see Table~\ref{table_vsigma}),
147: with that of G1 perhaps being slightly larger.
148: Since the reddening of M31~037--B327 is large and spatially
149: variable across the face of the cluster, we prefer to
150: compare the luminosities at $K$ where the impact of the reddening
151: is minimized. We assume both objects are
152: old stellar clusters.
153: This comparison is sure to be more reliable
154: than a similar one making the same assumption
155: carried out at $V$ by \cite{ma06a} which
156: suggested that M31~037--B327 is a factor of $\sim$2.5 times
157: more luminous than G1.
158:
159: In late September 2006 we acquired
160: an image at $K$ of the field of M31~037--B327
161: with the Wide Field Infrared Camera \citep{wilson03} at the Hale Telescope
162: on Palomar Mountain to verify the IR photometry of this
163: cluster. We find $K_s = 11.06\pm0.05$~mag
164: in the 2MASS \citep{2mass1,2mass2} system for an aperture
165: $11^{\arcsec}$ in diameter. This
166: is within 0.01~mag of that
167: derived from the 2MASS database
168: by \cite{galletti04}\footnote{Table~2 of \cite{galletti04} presents
169: IR mags measured from 2MASS transformed into the CIT
170: system using the transformation equation of
171: \cite{carpenter01}. We made the required inverse transformation
172: for the comparison quoted above.}.
173: $K_s$ for G1
174: from 2MASS as reported by
175: \cite{galletti04} is 0.025 mag fainter than that
176: of M31~037--B327.
177: If one takes $E(B-V)$ for the heavily reddened
178: cluster M31~037--B327
179: as $1.3$~mag with a more realistic error than that adopted by
180: \cite{ma06a},
181: given the spatially varying reddening, of $\pm 0.3$~mag,
182: then the M31~037--B327 has $M_K$ brighter by 0.16$\pm 0.03$ mag.
183: We thus find that M31~037--B327 has
184: $L_K$ comparable to that of G1 and at most 20\% larger, even
185: allowing for a generous
186: uncertainty in $E(B-V)$.
187: $M_K$ is well known to be a good measure of the
188: total luminosity for old stellar
189: systems; the dependence of $L_{\lambda}$ on [Fe/H] is smaller at 2.4$\mu$
190: than at optical wavelengths.
191:
192: In order to determine the virial mass of each of these objects,
193: we must assume that the same initial mass function prevails
194: in both of these clusters, and combine
195: $\sigma_v$ with the half light radius determined from HST
196: or other high spatial resolution imaging. \cite{ma06b} have
197: measured $R_h$ for M31~037--B327, while there are two independent
198: and discrepant determinations of $R_h$ for G1 (that of Meylen \etal\ 2001
199: and of Barmby, Holland \& Huchra 2002). $R_h$ for the highly reddened cluster
200: M31~037--B327 is not larger than that of G1.
201:
202: Thus there is no evidence that the highly reddened globular cluster
203: M31~037--B327 is substantially more massive than M31~G1.
204:
205:
206:
207: \section{Conclusions}
208:
209: The luminous object M31~037--B327 believed to be a globular
210: cluster in M31 is indeed a massive
211: object which may or may not be
212: the nucleus of an accreted galaxy, as was suggested by \cite{meylan01} and others.
213: However,
214: although it is among the most massive globular
215: clusters in M31, at least four other globular clusters, G1, G78, G280 and G213,
216: studied by \cite{george}, whose earlier $\sigma_v$ for G1 is
217: confirmed here,
218: have comparable masses. The first three of these
219: are probably more massive than M31~037--B327. Among the Galactic globular
220: clusters, $\omega$ Cen and NGC~6441 have $\sigma_v$ comparable
221: to that of M31~037--B327, based on the compilation
222: of \cite{pryor}. Contrary to the conclusion of \cite{ma06a}, we find
223: it to be very similar to the well studied cluster M31 G1.
224: M31~037--B327 is not an extraordinarily massive old cluster, and
225: is almost certainly not the most massive
226: cluster of any age in the Local Group.
227:
228:
229: \acknowledgements
230:
231: The entire Keck/HIRES user communities owes a huge debt to
232: Jerry Nelson, Gerry Smith, Steve Vogt, and many other
233: people who have worked to make the Keck Telescope and HIRES
234: a reality and to operate and maintain the Keck Observatory.
235: We are grateful to the W. M. Keck Foundation for the vision to fund
236: the construction of the W. M. Keck Observatory.
237: The authors wish to extend special thanks to those of Hawaiian ancestry
238: on whose sacred mountain we are privileged to be guests.
239: Without their generous hospitality, none of the observations presented
240: herein would have been possible.
241: This publication makes use of data from the Two Micron All-Sky Survey,
242: which is a joint project of the University of Massachusetts and the
243: Infrared Processing and Analysis Center, funded by the
244: National Aeronautics and Space Administration and the
245: National Science Foundation.
246: J.G.C. is grateful to NSF grant AST-0507219 for partial support.
247:
248: % REFERENCES
249:
250: \clearpage
251:
252:
253: \begin{thebibliography}{}
254:
255: \bibitem[Barmby \etal(2000)]
256: {barmby00}
257: Barmby, P., Huchra, J.~P., Brodie, J.~P., Forbes, D.~A.,
258: Schroder, L.~L. \& Grillmair, C.~J., 2000, \aj, 119, 747
259: % this the Barmby catalog of M31 GCs
260:
261: \bibitem[Barmby, Perrett \& Bridges(2002)]{barmby02a}
262: Barmby, P., Perrett, K.~M. \& Bridges, T.~J., 2002,
263: \mnras, 329, 461
264:
265: \bibitem[Barmby, Holland \& Huchra(2002)]{barmby02b}
266: Barmby, P., Holland, S. \& Huchra, J.~P., 2002, \aj,
267: 123, 1937
268:
269: \bibitem[Baumgardt \etal(2003)]{baumgardt}
270: Baumgardt, H., Makino, J., Hut, P., McMillan, S. \&
271: Portegies Zwart, S., 2003, \apjl, 589, L35
272:
273: \bibitem[Carpenter(2001)]{carpenter01}
274: Carpenter, J.~C., 2001, \aj, 121, 2851
275:
276: \bibitem[Cutri \etal(2003)]{2mass2} Cutri, R.~M. \etal, 2003,
277: ``Explanatory Supplement to the 2MASS All-Sky Data Release,
278: http://www.ipac.caltech.edu/2mass/releases/allsky/doc/explsup.html
279:
280: \bibitem[Djorgovski \etal(1997)]{george}
281: Djorgovski, S.~G. \etal, 1997, \apjl, 474, L19
282:
283: \bibitem[Galleti \etal(2004)]{galletti04}
284: Galleti S., F.~L., Bellazzini, M., Fusi Pecci, F. \& Macrina, S., 2004,
285: \aap, 416, 917
286:
287: \bibitem[Gebhardt, Rich \& Ho(2003)]{gebhardt}
288: Gebhardt, K., Rich, R.~M. \& Ho, L., 2002, \apjl, 578, L41
289:
290: \bibitem[Harris(1996)]{harris96} Harris, W., 1996, \aj, 112, 1487
291:
292: \bibitem[Ma \etal(2006a)]{ma06a}
293: Ma, J. \etal, 2006a, \mnras, 368, 1443
294:
295: \bibitem[Ma \etal(2006b)]{ma06b}
296: Ma, J. \etal, 2006b, \apjl, 636, L93
297:
298: \bibitem[Meylan \etal(2001)]{meylan01}
299: Meylan, G., Sarajedini, A., Jablonka, P., Djorgovski, S.~G.,
300: Bridges, T. \& Rich, R.~M., 2001, \aj, 122, 30
301:
302:
303: \bibitem [Pryor \& Meylan(1993)]{pryor}
304: Pryor, C. \& Meylan, G., 1993, in {\it{Structure and Evolution
305: of Globular Clusters} }, ed. S.Djorgovski \& G. Meylan,
306: ASPCS, 50, 357
307:
308: \bibitem [Sargent, Schechter, Boksenberg \& Sortridge(1977)]{sargent}
309: Sargent, W.~L.~W., Schechter, P.~L., Boksenberg, A. \&
310: Shortridge, K., 1977, \apj, 212, 326
311:
312: \bibitem[Skrutskie \etal(2006)]{2mass1}
313: Skrutskie, M.~F. \etal, 2006, \aj, 131, 1163
314:
315: \bibitem[Vogt \etal(1994)]{vogt94} Vogt, S.~E. \etal\, 1994, SPIE, 2198, 362
316:
317: \bibitem[Wilson \etal(2003)]{wilson03}
318: Wilson, W.~C. \etal, 2003, SPIE, 484, 4157
319: % WIRC design paper
320:
321: \end{thebibliography}
322:
323: \clearpage
324:
325: \begin{deluxetable}{l rrrr}
326: \tablenum{1}
327: \tablewidth{0pt}
328: \small
329: \tablecaption{Velocity Dispersion Measurements for 2 M31 Globular Clusters
330: \label{table_vsigma}}
331: \tablehead{ \colhead{ID} & \colhead{Wavelength Range} &
332: \colhead{$\sigma_v$} & \colhead{Uncertainty}
333: & \colhead{Ap.Cor. $\sigma_v$\tablenotemark{a}}\\
334: \colhead{} & \colhead{(\AA)} & \colhead{(1$\sigma$, \kms)} &
335: \colhead{(\kms)} & \colhead{(\kms)}
336: }
337: \startdata
338: M31~037--B327 & 5150--5190 & 19.2 & 3.5 & 21.9 \\
339: ~~~ & 6105--6190 & \nodata\tablenotemark{b} & \nodata & \nodata \\
340: ~ & 6545--6600 & 19.9 & 3.4 & 22.6 \\
341: ~~ \\
342: M31~G1 & 5150--5190 & 19.6 & 1.8 & 22.4 \\
343: ~~~ & 6105--6190 & 22.3 & 1.5 & 25.4 \\
344: ~~~ & 6545--6600 & 22.4 & 1.5 & 25.6 \\
345: \enddata
346: \tablenotetext{a}{Aperture correction set to a factor of 1.14
347: following \cite{george}.}
348: \tablenotetext{b}{Features were too weak to permit a reliable
349: determination of $\sigma_v$ in this spectral region.}
350: \end{deluxetable}
351:
352: \clearpage
353:
354: \begin{figure}
355: \epsscale{0.8}
356: \plotone{f1.ps}
357: \caption{The region of the spectrum including H$\alpha$
358: % the Mg~triplet
359: is shown for M31~037--B327 (top panel), M31~G1 (middle panel),
360: and a template luminous red giant. The spectra have been
361: normalized, shifted to the rest frame, and slightly smoothed
362: by a Gaussian with FWHM = 5~\kms.
363: The spectral features marked by arrows in the bottom panel are
364: all Fe~I lines.
365: \label{figure_vsigma}}
366: \end{figure}
367:
368: \end{document}
369:
370:
371: figure: f1.ps = /scr/jlc/m31_hires/hires_sep2006/m31b327_3plot.ps
372: generated from program plot_3spec.f
373: