1: %\documentclass[12pt,preprint]{aastex}
2: \documentclass{emulateapj}
3: %\usepackage{psfig}
4: \usepackage{apjfonts,times}
5: %\usepackage[figuresright]{rotating}
6:
7: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
8: % AUTHOR'S MACROS
9: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
10: \newcommand{\beq}{\begin{equation}}
11: \newcommand{\eeq}{\end{equation}}
12: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
13: % AUTHOR'S SYMBOLS
14: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
15: \newcommand{\gapprox}{$\stackrel {>}{_{\sim}}$}
16: \newcommand{\lapprox}{$\stackrel {<}{_{\sim}}$}
17: \def\alp{\mbox{$\alpha$}}
18: \def\farcd{\hbox{$.\mkern-4mu^\circ$}}
19: \def\farcm{\hbox{$.\mkern-4mu^\prime$}}
20: \def\farcs{\hbox{$.\!\!^{\prime\prime}$}}
21: \def\earth{\hbox{$\oplus$}}
22: \def\arcmin{\hbox{$^\prime$}}
23: \def\arcsec{\hbox{$^{\prime\prime}$}}
24: \def\solar{\mbox{$_{\normalsize\odot}$}}
25: \def\fs{\hbox{$.\!\!^{\rm s}$}}
26: \def\deg{\hbox{$^\circ$}}
27: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
28: \newcommand{\ttbs}{\char'134}
29: \newcommand{\AmS}{{\protect\the\textfont2
30: A\kern-.1667em\lower.5ex\hbox{M}\kern-.125emS}}
31: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
32: \newcommand{\lsim}{\ \raise
33: -2.truept\hbox{\rlap{\hbox{$\sim$}}\raise5.truept\hbox{$<$}\ }}
34: \newcommand{\gsim}{\ \rais-2.truept\hbox{\rlap{\hbox{$\sim$}}\raise5.truept\hbox{$>$}\ }}
35: \newcommand{\simsim}{\ \raise
36: -2.truept\hbox{\rlap{\hbox{$\sim$}}\raise5.truept\hbox{$\sim$}\ }}
37: \newcommand{\FeH}{\mbox{[Fe/H]}\,}
38: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
39:
40: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
41: % add words to TeX's hyphenation exception list
42: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
43: \hyphenation{author another re-commend-ed Post-Script}
44:
45: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
46: %% You can insert a short comment on the title page using the command
47: %% below.
48: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
49: \slugcomment{Accepted for Publication in ApJ --- Draft Version April
50: 22, 2007}
51:
52: %\righthead{}
53: %\lefthead{Gouliermis D., et al.}
54:
55: \shorttitle{HST/ACS observations of the intermediate-age cluster BS 90
56: in the SMC}
57: \shortauthors{Rochau B., et al.}
58:
59:
60: \begin{document}
61:
62: \title{The Star-forming Region NGC 346 in the Small Magellanic Cloud with
63: Hubble Space Telescope ACS Observations. II. Photometric Study of the
64: Intermediate-Age Star Cluster BS 90\footnote{Research supported by the
65: Deutsche Forschungsgemeinschaft (German Research Foundation)}
66: }
67:
68: %\footnotetext[]{Research supported by the Deutsche
69: %Forschungsgemeinschaft (German Research Foundation)}
70: %\altaffiltext{1}{Research supported by the Deutsche
71: %Forschungsgemeinschaft (German Research Foundation)}
72:
73:
74: \author{Boyke Rochau and Dimitrios A. Gouliermis}
75: \affil{Max Planck Institute for Astronomy, K\"onigstuhl 17, 69117
76: Heidelberg, Germany}
77: \email{rochau@mpia.de, dgoulier@mpia.de}
78:
79: \author{Wolfgang Brandner}
80: \affil{UCLA, Div. of Astronomy, 475 Portola Plaza, Los Angeles, CA
81: 90095-1547, USA}
82: \affil{Max Planck Institute for Astronomy, K\"onigstuhl 17, 69117
83: Heidelberg, Germany}
84: \email{brandner@astro.ucla.edu, brandner@mpia.de}
85:
86: \author{Andrew E. Dolphin}
87: \affil{Steward Observatory, University of Arizona, Tucson, AZ 85721, USA}
88: \affil{Raytheon Corporation, USA}
89: \email{adolphin@raytheon.com}
90:
91: \and
92:
93: \author{Thomas Henning}
94: \affil{Max Planck Institute for Astronomy, K\"onigstuhl 17, 69117
95: Heidelberg, Germany}
96: \email{henning@mpia.de}
97:
98: \begin{abstract}
99:
100: We present the results of our investigation of the intermediate-age star
101: cluster BS~90, located in the vicinity of the {\sc H ii} region N 66 in
102: the SMC, observed with HST/ACS. The high-resolution data provide a unique
103: opportunity for a very detailed photometric study performed on one of the
104: rare intermediate-age rich SMC clusters. The complete set of observations
105: is centered on the association NGC 346 and contains almost 100,000 stars
106: down to $V \simeq\ 28$ mag. In this study we focus on the northern part of
107: the region, which covers almost the whole stellar content of BS 90. We
108: construct its stellar surface density profile and derive structural
109: parameters. Isochrone fits on the CMD of the cluster results in an age of
110: about $4.5$ Gyr. The luminosity function is constructed and the
111: present-day mass function of BS 90 has been obtained using the
112: mass-luminosity relation, derived from the isochrone models. We found a
113: slope between $-1.30$ and $-0.95$, comparable or somewhat shallower than a
114: typical Salpeter IMF. Examination of the radial dependence of the mass
115: function shows a steeper slope at larger radial distances, indicating mass
116: segregation in the cluster. The derived half-mass relaxation time of
117: $0.95$ Gyr suggests that the cluster is mass segregated due to its
118: dynamical evolution. From the isochrone model fits we derive a metallicity
119: for BS 90 of $\rm{[Fe/H]}=-0.72$, which adds an important point to the
120: age-metallicity relation of the SMC. We discuss our findings on this
121: relation in comparison to other SMC clusters.
122:
123: \end{abstract}
124:
125: \keywords{Hertzsprung-Russell diagram --- Magellanic Clouds --- stars:
126: evolution --- clusters: individual ([BS95] 90)}
127:
128: \section{Introduction}\label{intro}
129:
130: Studies of star clusters, covering a wide range of ages, metallicities
131: and environments offer the opportunity to investigate the evolution of
132: individual clusters, but also of the entire parent galaxy, e.g. its star
133: formation history and chemical enrichment. The clusters of the Large
134: and Small Magellanic Cloud (LMC, SMC) are excellent targets for such studies.
135: Their proximity allows to resolve individual members of a stellar system
136: and to assume that all stars of a star cluster are more or less located
137: at the same distance. Studies of cluster systems of both the LMC and SMC
138: provide the opportunity to test theories of stellar evolution for
139: different ages, abundances and/or mass content in an environment
140: different than the Milky Way. Star clusters of the Magellanic Clouds
141: (MCs) have ages varying from the early stages of the clouds up to the
142: present, therefore allowing the study of the evolution of the clouds
143: from the time they have been formed until today.
144:
145: The history of cluster formation in the SMC is the topic
146: of ongoing discussion. Although the cluster population has been
147: considered as having a continuous age distribution (Da Costa \&
148: Hatzidimitriou 1998; Mighell et al. 1998), some studies suggest that most
149: of the SMC clusters have been formed in two epochs. Rich et al. (2000)
150: found seven populous SMC star clusters being formed either around $8\pm
151: 2$ Gyr or around $2\pm 0.5$ Gyr ago. Rafelski \& Zaritsky (2005) further
152: argue that the cluster age distribution shows few peaks, but no evidence
153: for a significant age gap in the SMC as observed in the LMC (van den Bergh 1991;
154: Da Costa 1991; Westerlund 1997). They suggest that the dominant
155: initial epoch of cluster formation together with a fast dissolution of
156: clusters makes the age distribution of the SMC clusters to appear with a
157: quiescent cluster formation period in the intermediate-age range.
158: Consequently, the intermediate-age range in the SMC is sparsely
159: populated.
160:
161: The low number of intermediate-age SMC clusters makes it difficult to
162: establish a consistent model for its chemical enrichment history. Pagel \&
163: Tautvai\v{s}ien\'{e} (1998) published two models to explain the enrichment
164: of the cluster population of both the SMC and LMC. Their work includes a
165: `bursting' model of star formation and a model with `smooth' star
166: formation. For the latter, a constant star formation rate is assumed over
167: the entire lifetime of the clouds. It is comparable to a model, which has
168: been previously published by Da Costa \& Hatzidimitriou (1998). The
169: bursting model assumes a constant star formation rate over a certain time
170: interval with a discontinuous change between two bursts. Pagel \&
171: Tautvai\v{s}ien\'{e} (1998) included two bursts for the SMC, one at the
172: time of its formation and the second at $\sim 2.5$ Gyr ago. Ages and
173: metallicities of the cluster population have been examined with
174: photometric analyses (e.g. Mighell et al. 1998; Piatti et al. 2001, 2005a)
175: or via spectroscopy (e.g. Da Costa \& Hatzidimitriou 1998; de Freitas
176: Pacheco et al. 1998; Piatti et al. 2005b). Results of this studies are not
177: providing a clear picture of the chemical enrichment of the SMC. As an
178: example, results from Mighell et al. (1998) are in better agreement with
179: the burst model of Pagel \& Tautvai\v{s}ien\'{e} (1998), whereas Da Costa
180: \& Hatzidimitriou (1998) pointed out that their findings are consistent
181: with a more continuous chemical evolution following a simple `closed-box'
182: model for the SMC, but they also mentioned that it may be an `open' rather
183: than a closed box due to its interaction with the LMC and the Galaxy.
184: Naturally, the further investigation of intermediate-age star clusters
185: would help to clarify the history of the cluster population of the SMC.
186:
187: Recent high-resolution imaging with the Advanced Camera for Surveys (ACS)
188: on-board the Hubble Space Telescope (HST) of the general area of the
189: stellar association NGC 346 in the SMC includes the intermediate-age star
190: cluster BS[95]~90 (Bica \& Schmitt 1995) or in short BS~90, providing,
191: thus, a unique opportunity to study one of the very rare star clusters,
192: which has been formed in a possible quiescent epoch of star formation in
193: the SMC. Such high-resolution studies are quite rare. The catalog of SMC
194: clusters imaged with the Wide-Field Planetary Camera 2 on-board HST,
195: presented by Mackey \& Gilmore (2003b), includes only 10 clusters, seven
196: of which have ages between 12 Gyr and 1.4 Gyr. The study we present here
197: on the cluster BS 90 is based on a unique set of deep observations toward
198: a SMC cluster with the high resolving power of HST/ACS. BS 90 is located
199: in the vicinity of the stellar association NGC 346, the largest stellar
200: concentration in the SMC, which is related to LHA 115-N 66 (in short N
201: 66), the brightest {\sc H ii} region of the galaxy (Henize 1956). NGC 346
202: has been subject of several studies. Peimbert et al. (2000) studied the
203: chemical composition of the region. High-sensitivity CO and H$_{\rm 2}$
204: observations and imaging in [{\sc O iii}] of the photodissociation region
205: is presented by Rubio et al. (2000). The X-ray emission in the general
206: area has been studied by Naz{\'e} et al. (2003, 2004). Finally, a snapshot
207: of the star formation history of the entire region is recently presented
208: by Sabbi et al. (2007).
209:
210: In the first paper of our investigation of the region of NGC 346 with
211: HST/ACS observations (Gouliermis et al. 2006, from here-on Paper I) we
212: present our photometry and the spatial distribution of different stellar
213: populations in the whole region, revealing the stellar richness of the
214: cluster BS 90. Our photometry allowed to resolve stars down to $V\simeq28$
215: mag ($\approx5$ mag below the turn-off), corresponding to $\sim0.4\
216: \rm{M_{\odot}}$. Naturally, the inclusion of BS 90 in this data set offers
217: the opportunity for a detailed analysis of this unique cluster using one
218: of the richest stellar samples in the SMC. Here, we present our results of
219: such an analysis. The structure of the paper is the following. In
220: \S~\ref{data} we discuss the data set and the data reduction. The
221: construction and analysis of the stellar surface density of the cluster is
222: presented in \S~\ref{structure}. In \S~\ref{cmdsection} the
223: color-magnitude diagram (CMD) of BS 90 is constructed and its properties
224: are derived by isochrone fitting. In \S~\ref{lfmf} the luminosity and
225: mass functions of the cluster are constructed and presented, and we
226: investigate the radial dependence of the mass function to examine a
227: possible mass segregation in \S~\ref{segregation}. We discuss our findings
228: concerning the age and metallicity of BS 90 in terms of the chemical
229: enrichment history of SMC in \S~\ref{ceh-smc}. Finally, in \S~\ref{sum} we
230: summarize our investigation of this unique SMC cluster.
231:
232: %\clearpage
233: %%%%%%%%%%%%%%%%%%%%%%%%%%%% FIGURE %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
234: \begin{figure}[t!]
235: \epsscale{1.25}
236: \plotone{f1.ps}
237: \caption{Completeness functions in $V$ ({\em left}) and $I$ ({\em
238: right}) for five different distances from the center of the cluster.
239: Annuli with radii of $0\farcm0-0\farcm2$, $0\farcm2-0\farcm4$,
240: $1\farcm0-1\farcm2$, $1\farcm8-2\farcm0$, $2\farcm6-2\farcm8$ have been
241: selected.}
242: \label{fig_compl}
243: \end{figure}
244: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
245:
246: \section{Data Description and Analysis}\label{data}
247:
248: The data we use in this study are collected within the HST Program
249: GO-10248. Three pointings were observed using the Wide-Field Channel
250: (WFC) of ACS, centered on the association NGC 346, in the broad-band
251: filters $F555W$ and $F814W$, equivalent to standard $V$ and $I$
252: respectively. The data sets are retrieved from the HST Data Archive.
253: Photometry of the pipeline-reduced FITS files was performed using the
254: ACS module of the package {\tt DOLPHOT} (version 1.0). This mode of the
255: package is an adaption of {\tt HSTphot} (Dolphin 2000), especially
256: designed for photometry on ACS imaging. With our photometry we detected
257: almost 100,000 stars. A detailed description of the datasets and a full
258: account of the photometric process is given in Paper I, where we also
259: made the whole photometric stellar catalog available.
260:
261: This catalog lists the celestial coordinates, and $V$- and
262: $I$-magnitudes (in the Vega system) with the corresponding photometric
263: errors for each star. Typical photometric uncertainties as a function
264: of the magnitude are shown in Paper I (Figure 2). The completeness of
265: the data has been calculated by running {\tt DOLPHOT} in artificial-star
266: mode as described in Paper I, and artificial star lists were created
267: with the utility {\em acsfakelist}. Completeness depends on the level of
268: crowding, and, consequently, in a populous compact cluster like BS 90 it
269: should vary with respect to the distance from the center of the cluster,
270: where crowding is higher. Indeed, the completeness in the area of BS 90
271: is found to be spatially variable. This is seen in Fig.~\ref{fig_compl},
272: where the completeness function in $V$ and $I$ (with the completeness
273: factors estimated as a function of magnitude as $N_{\rm detected}/N_{\rm
274: simulated}$) is shown for different selected distances from the center
275: of the cluster.
276:
277: \section{Dynamical Behavior of the Cluster}\label{structure}
278:
279: \subsection{Stellar Surface Density Profile of BS 90}\label{ssdp}
280:
281: The observed field covers a quite large area around BS 90, allowing the
282: construction of the stellar surface density profile of the cluster at relatively
283: large distances from its center. We divided the area of the cluster in
284: 30 concentric annuli with steps of $6\arcsec$ and we counted the number of stars
285: within each annulus. We corrected the counted stellar numbers for
286: incompleteness according to the corresponding completeness factors.
287: Completeness is a function of both distance from the center of the
288: cluster and magnitude. If $N_{i}(V)$ is the number of detected stars in
289: the $V$-band within the $i^{th}$ annulus and in the brightness range
290: corresponding to the completeness factor $c_{i}(V)$, the completeness
291: corrected number of stars is \begin{equation} N_{i,c}=\sum_{V}^{}
292: c_{i}(V)\cdot N_{i}(V). \end{equation} We obtained the stellar surface
293: density, $f_{i}$, by normalizing this number to the area of the
294: corresponding annulus as $f_{i}=N_{i,c}/A_{i}$, with $A_{i}$ being the
295: area of the $i^{th}$ annulus. The observed field-of-view does not fully
296: cover the northern and western extend of the cluster, and, therefore, for
297: the annuli not completely observed, we considered only the available area
298: for the estimation of the corresponding surface.
299:
300: The constructed stellar surface density profile is shown in
301: Fig.~\ref{fig-dpl}. The errors correspond to uncertainties due to the
302: counting process and they represent the Poisson statistics. As expected,
303: this density profile shows a smooth drop as a function of distance from
304: the center of the cluster. It drops to a uniform level, with the
305: exception of the small increase at the distance around $2\arcmin$. We
306: identify this increase as a small young cluster located east of BS 90.
307: The horizontal uniform level represents the stellar density of the field
308: of the galaxy. Its value is shown as the solid line in
309: Fig.~\ref{fig-dpl} and has been measured by fitting the models of Elson,
310: Fall \& Freeman (1987) to the stellar surface density profile as shown
311: below.
312:
313: %\clearpage
314: %%%%%%%%%%%%%%%%%%%%%%%%%%%% FIGURE %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
315: \begin{figure}[t!]
316: \epsscale{1.2}
317: \plotone{f2.ps}
318: \caption{Radial stellar surface density profile. The stellar numbers
319: have been corrected for incompleteness. The solid line shows the uniform
320: level of the stellar background density, which corresponds to a value of
321: $\sim0.55~{\rm stars}/{\Box \arcsec}$. The errors represent the
322: uncertainties due to Poisson statistics. The small density peak at
323: distance $\sim 120\arcsec$ is due to a young compact cluster to the east
324: of BS 90.}
325: \label{fig-dpl}
326: \end{figure}
327: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
328:
329: \subsection{Structural Parameters}\label{parameters}
330:
331: We applied both the empirical model by King (1962) and the model of
332: Elson, Fall \& Freeman (1987; from here-on EFF) to the stellar surface
333: density profile of BS 90 in order to obtain its structural parameters.
334: The EFF model is suited for clusters without tidal truncation, while
335: King's empirical model represents a tidally truncated cluster. Both
336: models provide the opportunity to derive accurate characteristic radii
337: for the cluster, such as its core and tidal radius. The core radius ($r_{\rm
338: c}$) describes the distance from the center of the cluster where the
339: stellar density drops to the half of its central value and the tidal radius
340: ($r_{\rm t}$) is the limit where the density drops to zero. For the
341: application of King's model the stellar surface density of the cluster
342: alone (with no contamination from the field) is necessary, while the EFF
343: model does not require any field subtraction. We first apply the latter
344: in order to estimate the core radius and the uniform background level.
345: The subtraction of this density level from the measured surface density
346: at each annulus gives the surface density profile of the cluster alone,
347: from which we derive the core and tidal radii.
348:
349: \subsubsection{Best Fitting EFF Profile}
350:
351: From studies of young LMC clusters by EFF it appears that the examined
352: clusters are not tidally truncated. These authors developed a model more
353: suitable to describe the stellar surface density profile of such
354: clusters:
355: \begin{equation}
356: f(r)=f_{0}(1+r^{2}/a^{2})^{-\gamma/2}+f_{\rm{field}}
357: \label{eq-elson}
358: \end{equation}
359: where $f_{0}$ is the central stellar surface density, $a$ is a measure
360: of the core radius and $\gamma$ is the power-law slope which describes
361: the decrease of surface density of the cluster at large radii
362: ($f(r)\propto r^{-\gamma/2}$ for $r\gg a$). $f_{\rm{field}}$ represents
363: the uniform background density level. Our best fit of the EFF model to
364: the surface density profile of BS 90 gives $a=8.24\pm0.67$ pc and
365: $\gamma=3.68\pm1.05$. The density of the background field, taken from the
366: fitting procedure, corresponds to a value of
367: $f_{\rm{field}}\sim0.55\pm0.04~{\rm stars}/{\Box \arcsec}$. According to
368: EFF model, the core radius, $r_{\rm c}$, is given from Eq.~\ref{eq-elson}
369: assuming no contribution from the field as:
370: \begin{equation}
371: r_{\rm c}=a(2^{2/\gamma}-1)^{1/2} \label{eq-rc}
372: \end{equation}
373: We estimate a core radius of $r_c=5.57^{+0.73}_{-0.37}$ pc. The
374: best-fitting EFF model assuming the derived values of $a, \gamma\
375: \rm{and}$ $f_{\rm{field}}$ is shown superimposed on the stellar surface
376: density profile in Fig.~\ref{fig-king}.
377:
378: %\clearpage
379: %%%%%%%%%%%%%%%%%%%%%%%%%%%% FIGURE %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
380: \begin{figure*}[t!]
381: \epsscale{1.}
382: \plotone{f3.ps}
383: \caption{Radial surface density profile of BS 90 with the best fitting
384: King and EFF models superimposed. The King model is overplotted with a
385: dashed line and the EFF model with a solid line. The error bars represent
386: Poisson statistics. {\em Left}: The profile with all detected stars,
387: including field and association. {\em Right}: The profile of the cluster
388: alone after subtracting the estimated background density level. Both
389: models fit very well the density profiles with a core radius of $r_{\rm
390: c}\simeq5.6$ pc (EFF) or $r_{\rm c}\simeq5.4$ pc (King), a power-law slope
391: of $\gamma=3.68$ (EFF), and a tidal radius of $r_{\rm t}\simeq53.7$ pc
392: (King).}
393: \label{fig-king}
394: \end{figure*}
395: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
396:
397: \subsubsection{Best fitting King Profile}\label{bestfitking}
398:
399: In order to have the density profile of the cluster alone we subtracted
400: the background density level of $\sim0.55\pm0.04~{\rm stars}/{\Box
401: \arcsec}$, estimated above, from the profile of Fig.~\ref{fig-dpl}. Then
402: we use the field-subtracted profile to derive the core and tidal radius
403: of the cluster as described by King (1962). Specifically according to this
404: model the core radius is found from the formula
405: \begin{equation}
406: f=\frac{f_0}{1+(r/r_{\rm c})^2}
407: \label{core}
408: \end{equation}
409: which describes the inner region of the cluster and the tidal radius from
410: \begin{equation}
411: f=f_1(1/r-1/r_{\rm t})^2
412: \label{tidal}
413: \end{equation}
414: which represents the outskirts of the cluster.
415: $f_0$ describes again the central surface density of the cluster and
416: $f_1$ is a constant. The best fitting profile results in a tidal radius
417: $r_{\rm t}=53.66^{+8.93}_{-7.10}$ pc
418: ($3\farcm13^{+0\farcm53}_{-0\farcm41}$) and a core radius $r_{\rm
419: c}=5.44^{+0.33}_{-0.35}$ pc ($19\farcs06^{+1\farcs15}_{-1\farcs23}$).
420: The concentration parameter, $c$, is defined as the logarithmic ratio of
421: tidal to core radius $c=\log{(r_{\rm t}/r_{\rm c})}$ and refers to the
422: compactness of the cluster. We obtained a concentration parameter
423: $c=1.0\pm0.1$. According to King's model the density profile of a
424: tidally truncated cluster is given as:
425: \begin{equation}
426: f\propto\Bigl(\frac{1}{[1+(\frac{r}{r_{\rm c}})^2]^\frac{1}{2}}-\frac{1}
427: {[1+(\frac{r_{\rm t}}{r_{\rm c}})^2]^\frac{1}{2}}\Bigr)^{2}
428: \label{eq-king}
429: \end{equation}
430: where $f$ is the stellar surface density, $r_{\rm c}$ and $r_{\rm t}$
431: the core and tidal radius, respectively and $r$ is the distance from the
432: center. The best fitting King model is shown as the dashed line,
433: superimposed on the radial surface density profile in
434: Fig.~\ref{fig-king}.
435:
436: Both King and EFF models are in good agreement to each other.
437: The intermediate region is equally well fitted by both models. Based on the
438: accuracy of the fits, it can be concluded that the EFF model is in better agreement
439: with our data in the inner part of the cluster but for the intermediate and
440: outer regions of the cluster the best fits yield no preference
441: for the tidally truncated King model or the EFF model.
442:
443: \subsubsection{Core Radius-Age Relation of SMC clusters}
444:
445: A comparison of 10 SMC clusters, investigated by Mackey \& Gilmore
446: (2003b), shows that they can be divided into two groups with respect to
447: their core radii, as they are derived from the EFF model. One group
448: contains six star clusters (NGC 176, 330, 121, 411, 416, 458) with radii
449: less than $5$ pc and the second group includes the remaining four clusters (NGC
450: 361, 152, 339, Kron 3) with $r_{\rm c}>5$ pc. BS 90 with its core
451: radius of $r_{\rm c}=5.57$ pc, as found from EFF models, would be a
452: member of the second group. Mackey \& Gilmore examined the relationship
453: between the core radii and the ages of clusters located in the LMC, SMC,
454: Fornax and Sagittarius dwarf spheroidals (see Mackey \& Gilmore
455: 2003a,b,c), and in Fig.~\ref{fig-core_age} the core radius-age
456: relationship for the SMC cluster sample is shown with our measurements
457: for BS 90. The distinction of the clusters into the two groups according
458: to their core radii is easily seen in this plot. For the LMC and the
459: Fornax and Sagittarius dwarf spheroidals they observed a similar
460: distinction and argued that it shows a real evolution of cluster
461: structure with age (Mackey \& Gilmore 2003a,c), with the sequence of
462: clusters with smaller core radii following the `standard' isolated
463: globular cluster evolution. These authors suggest that the evolution of
464: clusters at the upper right of Fig.~\ref{fig-core_age}, where BS 90
465: belongs to, differ from this isolated evolution. The disparity could be due
466: to very different stellar populations in these clusters or because they
467: do not follow an isolated evolution, but rather an evolution influenced
468: by external processes.
469:
470: %\clearpage
471: %%%%%%%%%%%%%%%%%%%%%%%%%%%% FIGURE %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
472: \begin{figure}[t!]
473: \epsscale{1.25}
474: \plotone{f4.ps}
475: \caption{Core radius versus age for the SMC clusters examined by Mackey
476: \& Gilmore (2003b) with the values for BS 90 found in this paper also
477: plotted. The triangles represent the clusters which have been
478: investigated by Mackey \& Gilmore (2003b). The star cluster BS 90 is
479: displayed as the black dot. Clusters which are located in the upper
480: right may have not been evolved in isolation.}
481: \label{fig-core_age}
482: \end{figure}
483: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
484:
485: McLaughlin \& van der Marel (2005) also derived core and tidal radii
486: with the application of King empirical models and the same separation of
487: the clusters into the two groups according to their core radii was
488: observed in their sample. Our application of King empirical models
489: yields a core radius of $r_{\rm c}\simeq5.4$ pc and a tidal radius of
490: $r_{\rm t}\simeq53.7$ pc, making BS 90 also a member of to the second group of the
491: McLaughlin \& van der Marel sample of clusters. Our estimated
492: concentration parameter for BS 90 $c\simeq1.0$ is relatively low in comparison to the SMC
493: clusters studied by Mackey \& Gilmore (2003b) and McLaughlin \& van der
494: Marel (2005).
495:
496: \subsubsection{Half-Light Radius}\label{halflight}
497:
498: The half-light radius of the cluster is derived from its surface
499: brightness profile. It corresponds more or less to the half-mass radius and
500: the corresponding half-mass relaxation time (\S~\ref{halfmass}) is a useful
501: reference time for the dynamical evolution of the cluster as a whole
502: (Spitzer 1987). To construct the surface brightness profile we subdivided the area of the
503: cluster into 26 annuli. The first 22 annuli were selected in steps of
504: $6\arcsec$ and the remaining four in steps of $12\arcsec$. We changed
505: the width of the annuli to improve the statistics in the outer regions
506: of the cluster. We converted the measured apparent magnitudes into flux
507: and counted the brightness of the stars within each annulus. The total
508: brightness of each annulus has been corrected for incompleteness and
509: normalized to the corresponding area. From our fit of the profile with
510: the EFF model we derived the total brightness of the entire cluster and
511: further obtained its half-light radius. We found a half-light radius of
512: $r_{\rm h}=9.42\pm0.5$ pc.
513:
514: \section{Color-Magnitude diagram}\label{cmdsection}
515:
516: The $V-I$ vs. $V$ color-magnitude diagram (CMD) of all stars detected
517: within a circular area of two core radii around the center of
518: the cluster is shown in Fig.~\ref{fig-cmd_all}. A comparison of the
519: contour maps of the region around NGC 346, constructed with star counts
520: in Paper I, shows that the main body of the cluster covers well the
521: northern part of the observed field-of-view (Paper I, Figure 4). Our
522: selected radius of 2 $\times$ $r_{\rm c}$ around the center of BS 90
523: corresponds to $11.15$ pc and is comparable to the half-light radius of
524: $9.42$ pc. It contains more than 40\% of the total number of observed stars and
525: more than 60\% of the observed cluster population. The CMD of
526: Fig.~\ref{fig-cmd_all} is indicative of an old cluster with a clear
527: turn-off at $V\sim22$ mag and $V-I\sim0.5$ mag, a well defined sub-giant
528: branch between $0.6 \lesssim V-I\lesssim1.0$ mag and $V\sim21.5$ mag, a
529: red giant branch between $1.0 \lesssim V-I\lesssim1.65$ mag and $17
530: \lesssim V\lesssim21.5$ mag and a populated red clump at $V\sim19.5$ mag
531: and $V-I\sim1.0$ mag. The CMD naturally includes also field stars and
532: stars of the nearby young association NGC 346 (see also Paper I). Part
533: of the latter is visible for instance as the population of the upper
534: main sequence.
535:
536: %\clearpage
537: %%%%%%%%%%%%%%%%%%%%%%%%%%%% FIGURE %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
538: \begin{figure}[t!]
539: \epsscale{1.25}
540: \plotone{f5.ps}
541: \caption{$V$, $V-I$ CMD of all detected stars within a circular area of
542: $2\ r_{\rm c}$, corresponding to $\sim 11.2$ pc, around the center of
543: the cluster BS 90. The turn-off is clearly visible and the features of
544: the red clump, the sub-giant branch and the red giant branch are easily
545: seen as well, making BS 90 a {\em bona-fide} case. This CMD is
546: contaminated by field stars and also by the close-by association NGC 346,
547: the population of which is apparent in the upper main sequence with
548: colors bluer $V-I\lesssim0.5$ mag.}
549: \label{fig-cmd_all}
550: \end{figure}
551: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
552:
553: %\clearpage
554: %%%%%%%%%%%%%%%%%%%%%%%%%%%% FIGURE %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
555: \begin{figure*}[t!]
556: \epsscale{1.}
557: \plotone{f6.ps}
558: \caption{{\em Left}: $V$, $V-I$ CMD of BS 90 after the contaminating
559: stellar population of the general SMC field and the association NGC 346
560: has been randomly subtracted with the Monte-Carlo method. The
561: contaminating stars of the upper main sequence, are almost completely
562: removed. Remaining stars with $21.5 \lesssim V\lesssim22.5$ mag and
563: $0.2\lesssim V-I\lesssim0.6$ mag may indicate a population of blue
564: stragglers. {\em Right}: The same CMD with the applied isochrone models
565: overplotted. The best-fitting isochrone, plotted by the middle thick
566: line, gives an age of $\tau\simeq4.5$ Gyr. The remaining two models,
567: plotted with thin lines, correspond to the limiting isochrones that can
568: also be fitted, and provide the error of 0.5 Gyr in the age estimation.
569: All three models fit the cluster best for a metallicity
570: $\rm{[Fe/H]}\simeq-0.72$, an assumed distance modulus of $m-M\simeq18.85$
571: mag, and a total visual extinction $A_V\sim0.04$ mag. The features of the
572: red clump, the red giant branch, the sub-giant branch and the turn-off are
573: all fitted accurately.}
574: \label{fig-cmd}
575: \end{figure*}
576: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
577:
578: \subsection{Subtraction of Contaminating Stars}\label{subtraction}
579:
580: In order to have a `clean' CMD, composed only by stars of the cluster,
581: a subtraction of all contaminating populations is necessary. Therefore
582: we applied a Monte Carlo subtraction method to clean the diagram from
583: stars, that belong to the field and the association.
584:
585: Naturally, this subtraction might depend on the area chosen to be field.
586: Considering that the observed field-of-view (FoV) is no more than
587: $5\arcmin \times 5\arcmin$ in size, and that the association NGC~346
588: itself also contaminates the stellar population of BS~90, we tested
589: several ``empty'' areas within the FoV, which also include stars from the
590: association, as potential fields, in order to determine the most realistic
591: stellar numbers that should be reduced from the CMD of BS~90. As a
592: reference for the stellar contamination we selected an area of
593: $75\arcsec\times50\arcsec$ in the south of BS 90 with its center
594: $\sim90\arcsec$ away from the center of the cluster. It contains stars of
595: the association, of the general field and also few stars of the cluster
596: itself. The number of stars that belong to the cluster is negligible at
597: distances of $r>65\arcsec$ from its center, because the stellar surface
598: density has almost dropped to the background level of the field at this
599: distance (see Fig.~\ref{fig-dpl}). We made the assumption that the
600: selected area contains only field stars and stars of the association and
601: not stars of BS 90. This leads to a small overestimation of the number of
602: stars to be subtracted and causes a slightly higher number of subtractions
603: in the area of the sub-giant branch and the main sequence.
604:
605: We divided the CMD into a grid and counted the numbers of stars in each
606: grid element of the CMD of the reference field. Considering that the
607: field-subtraction might also depend on the size of the elements, we tested
608: the Monte Carlo subtraction for different grid element sizes, and we found
609: that a size of 0.125 mag is the most suitable. Smaller element sizes
610: produce poor number statistics, while larger sizes result to rough
611: field-subtracted CMDs. Furthermore, we considered the different
612: completeness of the two regions by correcting the corresponding stellar
613: numbers for every field. Finally, we normalized the stellar numbers to the
614: selected area of the cluster. The resulting number of stars which
615: contaminates the CMD of BS 90 has been removed. The stars which have been
616: excluded have been randomly selected from the corresponding bin of the
617: original CMD of Fig.~\ref{fig-cmd_all}. The derived `clean' CMD of BS 90,
618: with no contamination by other stellar populations is shown in
619: Fig.~\ref{fig-cmd}.
620:
621: \subsection{The CMD of the Cluster}\label{cmdcluster}
622:
623: The CMD of Fig.~\ref{fig-cmd} ({\em left}), which corresponds to the CMD
624: of the star cluster BS 90 alone shows features of a {\em bona-fide}
625: intermediate-age cluster. The diagram reveals a well populated faint main
626: sequence, as well as a clear turn-off. Well-defined features are also the
627: sub-giant branch (SGB), the red giant branch (RGB) and the red clump (RC).
628: The asymptotic giant branch is sparsely populated. These clear
629: characteristics of the cluster allow isochrones to be easily used in order
630: to define several properties of BS 90. Specifically the SGB, the RGB and
631: the RC show no significant differential reddening, which makes the
632: application straightforward. We used isochrone models calculated by the
633: Padova Group (Girardi et al. 2002) in the VEGA filter system for ACS
634: filters.
635:
636: The best fitting isochrones are shown in the right panel of
637: Fig.~\ref{fig-cmd} superimposed on the CMD. The chosen isochrone fits
638: very well the turn-off, RC, SGB and RGB. These well-defined features of
639: BS 90 allow us to make an accurate estimation of the characteristic properties
640: of the cluster:
641:
642: \begin{itemize}
643: \item[{\em (i)}] {\em Age and Metallicity}\\
644: The best-fitting isochrones from the used model grid give an age of
645: $4.5\pm0.5$ Gyr and a metallicity of $Z=0.004\pm0.001$. This metallicity
646: leads through the relation ${\rm [Fe/H]}= 1.024\cdot \log{Z}+1.739$
647: (Bertelli et al. 1994) to an iron content for BS 90 of
648: $\rm{[Fe/H]}=-0.72^{+0.10}_{-0.13}$.
649:
650: \item[{\em (ii)}] {\em Reddening and Distance}\\
651: From our isochrone fit we found a color excess of only $E(V-I)\simeq
652: 0.016$ mag. Using the relation between the extinction in $V$, $A_V$, and
653: the color excess $E(V-I)$, derived by Mackey \& Gilmore (2003c), we
654: obtained a total visual extinction of $A_V\simeq 0.038$ mag. We assumed
655: a normal extinction law with a total-to-selective absorption of
656: $R_V=3.1$ (Binney \& Merrifield 1998), which gives $E(B-V)\simeq 0.021$
657: mag.
658:
659: The distance modulus of BS 90 has been determined to
660: $m-M\simeq18.85\pm0.1$ mag, which corresponds to a distance of
661: $58.9\pm0.45$ kpc.
662:
663: \end{itemize}
664:
665: %\clearpage
666: %%%%%%%%%%%%%%%%%%%%%%%%%%%% FIGURE %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
667: \begin{figure*}[t!]
668: \epsscale{1.}
669: \plotone{f7.ps}
670: \caption{The field-subtracted Luminosity Functions in $V$ ({\em left})
671: and $I$ ({\em right}) of the cluster within a radius $2\farcm8$. The
672: features of the red clump around $V\sim 19$ mag and $I\sim 18.5$ mag,
673: and the turn-off at $V\sim 21.5$ mag and $I\sim 21$ mag are clearly
674: seen. The increase of the stellar density for $V\gtrsim 26.5$ mag is
675: probably due to the very low completeness in our data. The vertical
676: dotted line represents the 70\% completeness limit, while the vertical
677: dash-dotted line represents the 50\% completeness limit. The errors
678: represent Poisson statistics. The dashed line shows the luminosity
679: function before it has been corrected for incompleteness.}
680: \label{fig-lf}
681: \end{figure*}
682: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
683:
684: \subsubsection{Blue Straggler Candidates}
685:
686: As shown in the CMD of Fig.~\ref{fig-cmd}, after the field subtraction
687: there are still few stars which remain located a bit above the turn-off
688: and to the blue. These stars appear to be main sequence stars with
689: masses larger than expected for stars of the cluster at its turn-off
690: age. The region of the CMD covered by these stars is between
691: $V\simeq22.5$ and $21$ mag, comparable to the region on the CMD, where
692: blue straggler stars are found in the LMC cluster ESO 121-SC03 (Mackey,
693: Payne \& Gilmore 2006). The spatial distribution of these candidate blue
694: stragglers in BS 90 shows that they are concentrated in the center of the
695: cluster. This supports the suggestion that these stars represent indeed a
696: blue straggler population, since such stars are centrally concentrated. A
697: suggested explanation for the Blue Stragglers phenomenon is mass
698: exchange in binary systems and merging processes, which increase the mass
699: of one star and, therefore, it shines brighter and is bluer than expected
700: for its age.
701:
702: \section{Luminosity Function and Mass Function}\label{lfmf}
703:
704: \subsection{Luminosity Function}\label{lf}
705:
706: We constructed the $V$- and $I$-band luminosity functions (LF) of the
707: cluster, which are presented in Fig.~\ref{fig-lf}. For the construction
708: of the LF we considered the CMD of the cluster shown in
709: Fig.~\ref{fig-cmd}, without the contamination by the populations of the
710: association and the background field within a radius $2\farcm8$ from the
711: center of the cluster (comparable to $r_{\rm t}$). We counted the stars
712: in magnitude bins of $0.25$ mag and corrected the numbers for
713: incompleteness. The LF has been normalized to the considered surface of
714: $6.5\ arcmin^2$. The completeness corrected LF is shown with thick
715: solid lines. The vertical dotted lines in Fig.~\ref{fig-lf}, which
716: correspond to $V=25.75$ mag and $I=24.5$ mag respectively, represent the
717: 50\% completeness limit. The vertical dashed-dotted line at $V=25$ mag
718: and $I=23.75$ mag respectively corresponds to the 70\% completeness
719: limit. The dashed line shows the LF before the completeness correction.
720:
721: In the LFs of Fig.~\ref{fig-lf} different features of the cluster can be
722: clearly identified. The red clump is visible as the increase in
723: stellar density around $V=19.5$ mag. The turn-off is also
724: distinguishable at $V\simeq 21.5$ mag. For fainter stars the LF turns
725: into a linearly increasing function, showing the main sequence stars of
726: the cluster. In the case of the $I$-LF we observe similar features.
727: Around $I=18.5$ mag we identify the red clump and we see the turn-off at
728: $I\simeq 20.75$ mag. For stars with $20.75 \lesssim I\lesssim22$ mag the
729: linear increase of the LF is steeper than for $22 \lesssim
730: I\lesssim24.5$ mag. The latter range is composed of stars of the main
731: sequence.
732:
733: \subsection{Mass Function}\label{mf}
734:
735: The IMF is a function which describes the mass distribution of a newly
736: formed stellar population. The stellar mass function
737: is changed under the influence of stellar, as well as dynamical evolution
738: that affects entire stellar systems such as star clusters (Baumgardt \& Makino 2003).
739: The present-day mass function (PDMF) is the result of such evolutionary effects.
740: In general, the IMF can be described as a power law (Scalo 1986):
741: \begin{equation}
742: \xi(\log{\rm m})\propto m^{\Gamma}
743: \label{imf}
744: \end{equation}
745: and represents the distribution of stellar masses for a given stellar system
746: at the time of their formation. The index $\Gamma$ is its power-law slope
747: \begin{equation}
748: \Gamma=\frac{d\log{\xi(\log{\rm m})}}{d\log{\rm m}}.
749: \label{gamma}
750: \end{equation}
751:
752: Salpeter (1955) found a value for the slope of the IMF in the solar
753: neighborhood, often taken as a reference, of $\Gamma=-1.35$. In the same
754: manner as for the IMF, we describe the slope of the PDMF as
755: \begin{equation}
756: a=\frac{d\log{N(\log{\rm m})}}{d\log{\rm m}}\ ,
757: \label{alpha}
758: \end{equation}
759: where $N(\log{\rm m})$ is the PDMF. This slope is given by the linear
760: relation between the logarithmic mass intervals and the corresponding
761: number of stars (in logarithmic scale). In order to obtain the PDMF of a
762: cluster, knowledge of the mass-luminosity relation (MLR) is required. In
763: the case of BS 90, and since a single age model fits almost perfectly
764: the cluster, we used the MLR taken from the 4.5 Gyr isochrone from
765: Girardi et al. (2002) and ascribed the present mass to each star which
766: belongs to our stellar sample of the cluster. To construct the PDMF of
767: BS 90, we counted stars in logarithmic mass intervals and normalized
768: their numbers to the considered area. We corrected these numbers for
769: incompleteness.
770:
771: The constructed PDMF is shown in Fig.~\ref{fig-mf}, where an almost
772: linear increase to lower masses can be seen. The low- and high-mass
773: ends of the PDMF show a steep increase and decrease respectively. In the
774: case of the low-mass end this is probably due to the completeness
775: corrections, while for the high-mass end stellar evolution is possibly
776: responsible. The 50\% completeness limit corresponds to a mass of
777: $M\sim0.7\ \rm{M_{\odot}}$ and is represented by the vertical
778: dash-dotted line. The 70\% completeness limit is shown with a vertical
779: dotted line and represents a lower mass limit of $M\sim0.8\
780: \rm{M_{\odot}}$. The straight solid line shows the fit to the mass
781: function above the 50\% completeness limit, without the two highest mass
782: points. The fit gives a single power-law slope of $a=-0.95\pm0.14$, a
783: bit flatter than Salpeter's IMF, whereas the fit to the mass function
784: above the 70\% completeness limit reveals a Salpeter-like slope of
785: $a=-1.30\pm0.14$. It is represented by the dashed line.
786:
787: Kroupa (2002) notes that theoretical examinations of the dynamics of massive
788: and long-lived globular clusters ($N\gtrsim 10^5$ stars) reveal that
789: the global PDMF is similar to the PDMF at distances near the half-mass
790: radius. The PDMF inward and outward the half-mass radius becomes
791: flatter and steeper, respectively, due to dynamical mass segregation.
792: Dynamical evolution of a cluster also flattens the global PDMF, since
793: evaporation is stronger for stars with lower masses than for high-mass
794: stars (Vesperini \& Heggie 1997).
795:
796: %\clearpage
797: %%%%%%%%%%%%%%%%%%%%%%%%%%%% FIGURE %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
798: \begin{figure}[t!]
799: \epsscale{1.2}
800: \plotone{f8.ps}
801: \caption{The field-subtracted, completeness corrected present-day mass
802: function of the cluster within a radial distance of $2\farcm8$ from its
803: center. The steep increase to the lower mass end ($\log{(M/{\rm
804: M}_{\odot})} <-0.25$) is probably due to the very low completeness. The
805: vertical dotted line represents the 70\% completeness limit while the
806: vertical dash-dotted line represents the 50\% completeness limit. The
807: best linear fit for data with completeness higher than 50\%,
808: corresponding to a slope of $a\simeq-0.95$ is plotted with a solid
809: line. The dashed line shows the linear fit to the mass function for
810: stars with completeness higher than 70 \% with a slope of $a\simeq-1.30$.}
811: \label{fig-mf}
812: \end{figure}
813: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
814:
815: \section{Mass segregation in BS 90}\label{segregation}
816:
817: The phenomenon of mass segregation describes the concentration of
818: high-mass stars towards the center of a star cluster. This phenomenon
819: could be either of primordial origin or the result of the dynamical
820: evolution of the cluster. The latter effect leads the cluster to a state of
821: energy equipartition between its stars. Therefore, stars with lower
822: masses will have higher velocities and thus larger orbits. On the other
823: hand, stars with higher masses have smaller orbits and
824: concentrate towards the center. Indeed, stellar mass functions (MFs) at
825: the outer parts of star clusters indicate the existence of more low-mass
826: stars at larger radii than in the central region of the cluster (de
827: Grijs et al. 2002). A powerful diagnostic for the detection of the
828: phenomenon of mass segregation in a star cluster is its MF, which is expected to be radially
829: variable if segregation is present (e.g. Gouliermis et al. 2004).
830: Therefore, in order to see if BS 90 is mass segregated or not, we derived
831: its MF for annuli at different radial distances from the center of the
832: cluster. We derived the different MFs as described in \S~\ref{mf} for
833: four selected radial distances $0\farcm0-0\farcm4$, $0\farcm4-0\farcm8$,
834: $0\farcm8-1\farcm2$ and $1\farcm2-1\farcm6$. The resulting MFs are shown
835: in Fig.~\ref{fig-mseg}. The MF of the innermost region has a slope of
836: $a=-0.48\pm0.14$. The MF of the second annulus, which includes the
837: half-light radius, is comparable to the MF inside $2\farcm8$ shown in
838: Fig.~\ref{fig-mf} and has a slope of $a=-0.96\pm0.12$. The slopes of
839: the outer regions are found to be steeper with $a=-2.06\pm0.58$ for
840: $0\farcm8-1\farcm2$ and $a=-1.96\pm0.31$ for $1\farcm2-2\farcm8$. The
841: difference of the MF slope being steeper outwards, gives clear evidence
842: of mass segregation in BS 90. In the following section, we will explore the
843: driving mechanism of this phenomenon.
844:
845: %\clearpage
846: %%%%%%%%%%%%%%%%%%%%%%%%%%%% FIGURE %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
847: \begin{figure*}[t!]
848: \epsscale{1.}
849: \plotone{f9.ps}
850: \caption{Mass functions of BS 90 for four selected radial distances
851: ($0\farcm0-0\farcm4$, $0\farcm4-0\farcm8$, $0\farcm8-1\farcm2$,
852: $1\farcm2-2\farcm8$). The dotted and dashed-dotted lines represent the
853: 70\% and 50\% completeness limit respectively. The linear fits to the
854: MFs are represented by the solid line in each panel. The mass function
855: becomes steeper for larger radial distances from the center of the
856: cluster indicating that mass segregation takes place.}
857: \label{fig-mseg}
858: \end{figure*}
859: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
860:
861: \subsection{Half-Mass Radius and Dynamical Time-Scale}\label{halfmass}
862:
863: As mentioned above, the phenomenon of mass segregation may be
864: primordial, due to the formation of the most massive stars in the
865: central regions of the cluster, or an evolutionary effect due to the
866: dynamical relaxation of the cluster. The latter is achieved via
867: two-body encounters, which lead to a quasi-Maxwellian equilibrium in
868: the interior of the cluster (Lightman \& Shapiro 1978). The tendency of
869: the inner regions of the cluster to reach thermal equilibrium leads to
870: energy equipartition among different stellar mass groups and
871: consequently to mass stratification. The time-scale in which such
872: equipartition is accomplished is given roughly by the relaxation time,
873: $t_{\rm rl}$, of the cluster (e.g. Spitzer 1975). Hence, the relaxation
874: time can indicate whether a case of mass segregation is of dynamical
875: origin (when the evolution time of the stellar system is longer than the
876: relaxation time) or not. This time-scale can be expressed as (Binney \&
877: Tremaine 1987):
878: \begin{equation}
879: t_{\rm rl}=\frac{6.63\times10^8}{\ln{0.4N}}\left(\frac{M}{10^5{\rm M}_{\odot}}\right)^{1/2}\left(\frac{{\rm M}_{\odot}}{m_{\star}}\right)\left(\frac{\mathit{R}}{\rm pc}\right)^{3/2}~~~{\rm yr}
880: \label{eq-rel}
881: \end{equation}
882: where $M$ is the total mass in a certain characteristic radius $R$,
883: $m_{\star}$ is a characteristic mass (median mass of the observed
884: stellar distribution) and $N$ is the corresponding number of stars of
885: the system. The half-mass radius corresponds more or less to the
886: half-light radius of the cluster. The half-light radius of BS 90 derived
887: from its surface brightness profile (\S~\ref{halflight}) is found to be equal
888: to $r_{\rm h}=9.42\pm0.5$ pc. Following Eq.~\ref{eq-rel}, we derive the
889: relaxation times, corresponding to the half-light and core radii, after
890: we estimated the corresponding total and characteristic mass and the
891: total number of stars within these radii. We found a half-mass
892: relaxation time of $t_{\rm rl,h}=0.95\pm0.06$ Gyr and a core radius
893: relaxation time of $t_{\rm rl,c}=0.33^{+0.06}_{-0.04}$ Gyr. From these
894: timescales, which are much smaller than the evolutionary age of the
895: cluster, we easily conclude that BS 90 is a dynamically relaxed
896: spherical cluster, which thus exhibits the phenomenon of dynamical mass
897: segregation.
898:
899: \section{Chemical Evolution History of the SMC}\label{ceh-smc}
900:
901: The chemical evolution history of a galaxy is understood in terms of the
902: age-metallicity relation of both field and cluster stars. The chemical
903: evolution histories in the MCs exhibit distinct features in comparison
904: to this of the disc of the Milky Way (Westerlund 1997). Several
905: models have been developed for their explanation, assuming bursts with
906: IMFs steeper than Salpeter's IMF to produce subsolar metallicities (e.g.
907: Russell \& Dopita 1992; Tsujimoto et al. 1995), or postulating selective
908: stellar outflows associated with star formation bursts (e.g. Pilyugin
909: 1996). In contrast, the model of chemical evolution for the MCs
910: developed by Pagel \& Tautvai\v{s}ien\'{e} (1998) assumes that the
911: Clouds have been built up by gradual infall of unprocessed material,
912: linear laws of star formation, yields and time delays identical to those
913: for the solar neighborhood, and a non-selective wind proportional to
914: the star formation rate. These models represent chemical enrichment
915: assuming a burst-like star formation (`bursting' model) and a continuous
916: star formation rate over the entire lifetime of the SMC (`smooth', or simple
917: `closed box' model).
918:
919: %\clearpage
920: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
921: \begin{table*}
922: \caption{Collective Data on ages and metallicities of star clusters in the SMC. \label{table1}}
923: \begin{center}
924: \scriptsize{
925: \begin{tabular}{lccrrrrrrc}
926: \hline
927: Name & RA & DEC & Age & & & Metal. & & & References\\
928: &\multicolumn{2}{c}{J2000}& (Gyr) & + & $-$ & [Fe/H] & + & $-$ & \\
929: \hline
930: \hline
931: NGC 121 & 00:26:49 & $-$71:32:10 & 12.00 & 2.00 & 2.00 & $-$1.19 & 0.12 & 0.12 & Da Costa \& Hatzidimitriou (1998)\\
932: & & & 11.90 & 1.30 & 1.30 & $-$1.71 & 0.10 & 0.10 & Mighell et al. (1998)\\
933: & & & 12.00 & 5.00 & 5.00 & $-$1.20 & 0.32 & 0.32 & de Freitas Pacheco et al. (1998)\\
934: NGC 152 & 00:32:56 & $-$73:06:57 & 1.40 & 0.20 & 0.20 & $-$0.94 & 0.15 & 0.15 & Crowl et al. (2001)\\
935: & & & 1.90 & 0.50 & 0.50 & $-$0.80 & 0.30 & 0.30 & Da Costa \& Hatzidimitriou (1998)\\
936: NGC 176 & 00:35:59 & $-$73:09:57 & 0.20 & 0.20 & 0.10 & $-$0.60 & - & - & Mackey \& Gilmore (2003b)\\
937: NGC 339 & 00:57:48 & $-$74:28:00 & 4.00 & 1.50 & 1.50 & $-$1.19 & 0.10 & 0.10 & Da Costa \& Hatzidimitriou (1998)\\
938: & & & 6.30 & 1.30 & 1.30 & $-$1.50 & 0.14 & 0.14 & Mighell et al. (1998)\\
939: & & & 2.00 & 0.70 & 0.70 & $-$0.70 & 0.22 & 0.22 & de Freitas Pacheco et al. (1998)\\
940: NGC 361 & 01:02:13 & $-$71:36:16 & 8.10 & 1.20 & 1.20 & $-$1.45 & 0.11 & 0.11 & Mighell et al. (1998)\\
941: NGC 411 & 01:07:56 & $-$71:46:05 & 1.40 & 0.20 & 0.20 & $-$0.68 & 0.07 & 0.07 & Alves \& Sarajedini (1999)\\
942: & & & 1.30 & 0.50 & 0.30 & $-$0.70 & 0.22 & 0.22 & de Freitas Pacheco et al. (1998)\\
943: & & & 1.50 & 0.30 & 0.30 & $-$0.70 & 0.20 & 0.20 & Piatti et al. (2005b)\\
944: & & & 1.80 & 0.30 & 0.30 & $-$0.84 & 0.30 & 0.30 & Da Costa \& Hatzidimitriou (1998)\\
945: NGC 416 & 01:07:59 & $-$72:21:20 & 6.90 & 1.10 & 1.10 & $-$1.44 & 0.12 & 0.12 & Mighell et al. (1998)\\
946: & & & 4.00 & 1.50 & 1.50 & $-$0.80 & 0.23 & 0.23 & de Freitas Pacheco et al. (1998)\\
947: NGC 419 & 01:08:19 & $-$72:53:03 & 4.00 & 1.50 & 1.50 & $-$0.60 & 0.21 & 0.21 & de Freitas Pacheco et al. (1998)\\
948: & & & 1.20 & 0.50 & 0.50 & $-$0.70 & 0.30 & 0.30 & Da Costa \& Hatzidimitriou (1998)\\
949: NGC 458 & 01:14:53 & $-$71:32:59 & 0.20 & 0.10 & 0.10 & $-$0.23 & 0.10 & 0.40 & Da Costa \& Hatzidimitriou (1998)\\
950: & & & 0.13 & 0.06 & 0.06 & $-$0.23 & 0.10 & 0.40 & Piatti et al. (2005b)\\
951: L1 & 00:04:00 & $-$73:28:00 & 10.00 & 2.00 & 2.00 & $-$1.01 & 0.11 & 0.11 & Da Costa \& Hatzidimitriou (1998)\\
952: & & & 9.00 & 1.00 & 1.00 & $-$1.35 & 0.08 & 0.08 & Mighell et al. (1998)\\
953: L4 & 00:21:27 & $-$73:44:55 & 3.10 & 0.90 & 0.90 & $-$0.90 & 0.20 & 0.20 & Piatti et al. (2005a)\\
954: L5 & 00:22:40 & $-$75:04:29 & 4.10 & 0.90 & 0.90 & $-$1.20 & 0.20 & 0.20 & Piatti et al. (2005a)\\
955: & & & 3.00 & 1.50 & 1.50 & $-$1.10 & 0.20 & 0.20 & Piatti et al. (2005b)\\
956: L6/K4 & 00:23:04 & $-$73:40:11 & 3.30 & 0.90 & 0.90 & $-$0.90 & 0.20 & 0.20 & Piatti et al. (2005a)\\
957: L7/K5 & 00:24:43 & $-$73:45:18 & 2.00 & 0.20 & 0.20 & $-$0.60 & 0.20 & 0.20 & Piatti et al. (2005a)\\
958: & & & 1.20 & 0.50 & 0.50 & $-$0.50 & 0.20 & 0.20 & Piatti et al. (2005b)\\
959: L11/K7 & 00:27:45 & $-$72:46:53 & 3.50 & 1.00 & 1.00 & $-$0.81 & 0.13 & 0.13 & Da Costa \& Hatzidimitriou (1998)\\
960: & & & 3.50 & 0.50 & 0.50 & $-$1.00 & - & - & Piatti et al. (2005b)\\
961: L19 & 00:37:42 & $-$73:54:30 & 2.10 & 0.20 & 0.20 & $-$0.75 & 0.20 & 0.20 & Piatti et al. (2005a)\\
962: L27 & 00:41:24 & $-$72:53:27 & 2.10 & 0.20 & 0.20 & $-$1.30 & 0.30 & 0.30 & Piatti et al. (2005a)\\
963: L32 & 00:47:24 & $-$68:55:10 & 4.80 & 0.50 & 0.50 & $-$1.20 & 0.20 & 0.20 & Piatti et al. (2001)\\
964: L38 & 00:48:50 & $-$69:52:11 & 6.00 & 0.60 & 0.60 & $-$1.65 & 0.20 & 0.20 & Piatti et al. (2001)\\
965: L113 & 01:49:30 & $-$73:43:40 & 6.00 & 1.00 & 1.00 & $-$1.17 & 0.12 & 0.12 & Da Costa \& Hatzidimitriou (1998)\\
966: & & & 5.30 & 1.30 & 1.30 & $-$1.24 & 0.11 & 0.11 & Mighell et al. (1998)\\
967: L116 & 01:55:33 & $-$77:39:16 & 2.80 & 1.00 & 1.00 & $-$1.10 & 0.20 & 0.20 & Piatti et al. (2001)\\
968: K3 & & & 6.00 & 1.30 & 1.30 & $-$1.16 & 0.09 & 0.09 & Mighell et al. (1998)\\
969: & & & 9.00 & 2.00 & 2.00 & $-$0.98 & 0.12 & 0.12 & Da Costa \& Hatzidimitriou (1998)\\
970: & & & 3.50 & 1.50 & 1.50 & $-$1.00 & 0.28 & 0.28 & de Freitas Pacheco et al. (1998)\\
971: & & & 7.00 & 1.00 & 1.00 & $-$1.20 & 0.20 & 0.20 & Piatti et al. (2005b)\\
972: K28 & 00:51:42 & $-$71:59:52 & 2.10 & 0.50 & 0.50 & $-$1.20 & 0.20 & 0.20 & Piatti et al. (2001)\\
973: & & & 1.50 & 0.60 & 0.60 & $-$1.00 & 0.20 & 0.20 & Piatti et al. (2005b)\\
974: K44 & 01:02:04 & $-$73:55:31 & 3.10 & 0.80 & 0.80 & $-$1.10 & 0.20 & 0.20 & Piatti et al. (2001)\\
975: HW 47 & 01:04:04 & $-$74:37:09 & 2.80 & 0.90 & 0.90 & $-$1.00 & 0.40 & 0.40 & Piatti et al. (2005a)\\
976: HW 84 & 01:41:28 & $-$71:09:58 & 2.40 & 0.20 & 0.20 & $-$1.20 & 0.40 & 0.40 & Piatti et al. (2005a)\\
977: HW 86 & 01:42:22 & $-$74:10:24 & 1.60 & 0.20 & 0.20 & $-$0.75 & 0.40 & 0.40 & Piatti et al. (2005a)\\
978: BS 121 & 01:04:22 & $-$72:50:52 & 2.30 & 0.20 & 0.20 & $-$1.20 & 0.40 & 0.40 & Piatti et al. (2005a)\\
979: BS90 & 00:59:05 & $-$72:09:10 & 4.30 & 0.10 & 0.10 & $-$0.84 & - & - & Sabbi et al. (2007)\\
980: & & & 4.47 & 0.55 & 0.49 & $-$0.71 & 0.10 & 0.13 & This Paper\\
981: \hline
982: \hline
983: \end{tabular}
984: }
985: \end{center}
986: \end{table*}
987: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
988:
989:
990: In order to reproduce the age-metallicity relation for individual clusters
991: in the SMC and add our results on BS 90, we collected a large set of data
992: available on SMC clusters, concerning their metallicities and ages. These
993: data are given in Table~\ref{table1}. We plot the age-metallicity relation
994: for the SMC clusters in Fig. ~\ref{fig-amr}. The models published by Pagel
995: \& Tautvai\v{s}ien\'{e} (1998) are also displayed with a solid (`bursting'
996: model) and a dash-dotted line (`smooth' model). The upper panel of
997: Fig.~\ref{fig-amr} shows ages and metallicities with the corresponding
998: errors for all star clusters collected in Table~\ref{table1}. The lower
999: panel is displayed without the errors, but with a set of dashed lines,
1000: which show the connections of points derived from different studies for
1001: the same clusters. The latter shows that both the `bursting' and the
1002: `smooth' model can be supported by selecting certain clusters and hence an
1003: unambiguous conclusion is not possible yet. Moreover, it should be noted
1004: that the data given in Table~\ref{table1} consist of an inhomogeneous set
1005: of abundances, perhaps washing away any real effects. The transformation of
1006: all these results into a common spectroscopic scale would certainly
1007: provide a more clear picture. In any case if we assume that the `bursting'
1008: model of Pagel \& Tautvai\v{s}ien\'{e} (1998) represents the chemical
1009: evolution history of SMC, then our measurements reveal a relatively high
1010: metallicity for the epoch during which BS 90 was formed. These values seem
1011: to fit better the `smooth' model, which represents a continuous mode of
1012: chemical enrichment.
1013:
1014: %\clearpage
1015: %%%%%%%%%%%%%%%%%%%%%%%%%%%% FIGURE %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1016: \begin{figure}[t!]
1017: \epsscale{1.25}
1018: \plotone{f10.ps}
1019: \caption{Age-metallicity relation of the cluster population in the SMC.
1020: The data are obtained from Crowl et al. (2001; small grey dot), Piatti
1021: et al. (2001, 2005a,b; small open circles), Mighell et al. (1998;
1022: diamonds), Da Costa \& Hatzidimitriou (1998; triangles), de Freitas
1023: Pacheco (1998; asterisks), Mackey \& Gilmore (2003b; small black dot),
1024: Alves \& Sarajedini (1999; square), Sabbi et al. (2007; large grey dot).
1025: Our measurements for BS 90 are presented as the large black dot. In
1026: both panels the bursting (solid curve) and the smooth (dotted-dashed
1027: curve) chemical evolution models by Pagel \& Tautvai\v{s}ien\'{e} (1998)
1028: are also shown. {\em Upper panel}: The values for all studied SMC
1029: clusters (shown in Table~\ref{table1}) are plotted with their
1030: corresponding errors. {\em Lower panel}: The same values from
1031: Table~\ref{table1} are shown without their errors. The dashed lines
1032: connect the values from estimates derived from different studies for the
1033: same clusters, exhibiting, thus, the discrepancy in the results of
1034: different methods.}
1035: \label{fig-amr}
1036: \end{figure}
1037: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1038: %Smooth model represents a continuous mode of enrichment is shown by the
1039: %dotted line.
1040: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1041:
1042: Indeed, our estimated age and metallicity for BS 90 places the cluster in
1043: Fig.~\ref{fig-amr} closer to the `smooth' model (large black dot), in
1044: agreement with the estimations of Sabbi et al. (2007) (large grey dot).
1045: According to the `bursting' model an iron abundance of
1046: [Fe/H]$\approx-1.2$ corresponds to the time around 4.5 Gyr ago. Our
1047: derived value for BS 90 is about $0.5$ dex higher. On the other hand the
1048: `smooth' model requires an iron content of [Fe/H]$\approx-0.6$ at $\sim
1049: 4.5$ Gyr, much closer to our derived value of [Fe/H]$\approx-0.7$. A
1050: simple `closed box' model is also supported by the results of Kayser et
1051: al. (2006) and consistent with the enrichment history of the outer SMC
1052: field which has been derived by Dolphin et al. (2001). With its age and
1053: metallicity BS 90 adds an important point to the age-metallicity
1054: relation, which favors a continuous mode of chemical enhancement in the
1055: SMC, but the chemical evolution history of this galaxy still remains
1056: elusive and seems to be more complicated than reflected by the
1057: proposed models. Fig.~\ref{fig-amr} also reveals an increasing number of
1058: clusters for the last $\sim 3$ Gyr. This coincides with the increase of
1059: cluster formation in the LMC and is consistent with a close encounter
1060: between the Magellanic Clouds and Milky Way as proposed by Bekki et al.
1061: (2004). The strong tidal interactions could certainly trigger cluster
1062: formation.
1063:
1064:
1065: \section{Summary}\label{sum}
1066:
1067: In this paper we present our detailed photometric analysis of the
1068: intermediate age spherical star cluster BS 90 in the SMC.
1069:
1070: We analyzed the stellar surface density profile of BS 90, following the
1071: models described by King (1962) and Elson, Fall \& Freeman (1987). We
1072: derived almost the same value for the core radius of $r_{\rm
1073: c}\simeq5.5$ pc with both models. This value places BS 90 in the $r_{\rm
1074: c}$-age relation among the SMC clusters that do not seem to evolve
1075: isolated. From the model of Elson, Fall \& Freeman (1987), we found a
1076: stellar surface density which, at larger radii, decreases outwards with
1077: a power-law slope of $\gamma\simeq3.68$. The application of King's model
1078: delivered a tidal radius of $r_{\rm t}\simeq53.7$ pc and a concentration
1079: parameter $c\simeq1.0$, which seems to be relatively low, compared to other
1080: SMC clusters (McLaughlin \& van der Marel 2005).
1081:
1082: We counted stars in different magnitude intervals in order to derive the
1083: LF of the cluster. The derived luminosity function
1084: of BS 90 reflects features of the cluster like the red clump as a higher
1085: stellar density around $V=19$ mag and $I=18.5$ mag, respectively, and the
1086: turn-off at $V=22$ mag and $I=20.75$ mag, respectively. The PDMF is a
1087: linearly decreasing function up to the high mass end of the cluster. We
1088: derived a slope of the PDMF of $a\simeq -0.95$ for a completeness in our
1089: photometry higher than 50 \% and of $a\simeq -1.30$ above the 70\%
1090: completeness limit. Consequently, the PDMF of BS 90 has a slope between
1091: the one derived by Salpeter (1955) for the IMF in the solar neighborhood,
1092: and a shallower slope, comparable to earlier derived slopes from studies
1093: on Magellanic Cloud clusters (e.g. Kerber \& Santiago 2006; Gouliermis
1094: et al. 2004). The difference from the Salpeter slope can also be explained
1095: by the fact, that unresolved binaries flatten the PDMF (Kerber \& Santiago
1096: 2006) or that stars escaped from the cluster (Meylan \& Heggie 1997).
1097:
1098: We investigated the radial dependence of the mass function slope to
1099: check whether BS 90 is mass segregated or not. We found that indeed the
1100: mass function becomes steeper at larger distances from the center of the
1101: cluster. The radial dependence of the slope indicates a central
1102: concentration of the massive stars and hence mass segregation. We also
1103: confirm previous claims that the PDMF around the half-mass radius is
1104: comparable to the overall PDMF since it is the region which is least
1105: affected by mass segregation (Kroupa 2002; Vesperini \& Heggie 1997).
1106: Within the half-light radius of $\simeq 9.4$ pc, we obtained a half-mass
1107: relaxation time of about $0.95$ Gyr and therefore we conclude that the
1108: mass segregation is due to the dynamical evolution of BS 90, since the
1109: age of the cluster is older than the derived half-mass relaxation time.
1110: Mass segregation provides another possible reason for a flattening of
1111: the mass function which may occur during the evolution of the cluster.
1112:
1113: Isochrone models fitting to the constructed CMD (Fig.~\ref{fig-cmd}) of
1114: BS 90 revealed an age of about $4.5$ Gyr, which characterizes this
1115: cluster as one of the very rare star clusters, formed in a possible
1116: quiescent period of cluster formation in the SMC (Rafelski \& Zaritsky 2005).
1117: There are only few clusters in the range between 4 and 10
1118: Gyr. The applied isochrone yields a metallicity of $\rm{[Fe/H]}=-0.72$.
1119: The total visual extinction has been found to be $A_{V}\sim 0.04$ mag. The
1120: low value of the total visual extinction combined with the large amount
1121: of gas in the general region and the absence of indications of
1122: differential reddening in the CMD of BS 90 suggests that the cluster is
1123: probably located in front of the association NGC 346. The distance of BS
1124: 90 has been estimated to $\simeq$ 58.9 kpc.
1125:
1126: The location of BS 90 in the age-metallicity plot shows that this cluster
1127: fits better to a simple `smooth' model, which represent a continuous
1128: chemical enhancement for the SMC, rather than a `bursting' model for the
1129: chemical evolution history of the galaxy, in agreement with recent results
1130: on other SMC clusters (Kayser et al. 2006). However, inconsistent
1131: metallicity estimation methods for different SMC clusters provide an
1132: inhomogeneous set of results, that should be transfered onto a common
1133: abundance scale, so that an unambiguous picture of the chemical evolution
1134: history of SMC can be derived. Furthermore, given that the SMC is part of an interacting system between the MCs and the Milky Way, the chemical evolution history of this galaxy is probably far more complicated than what suggested by a simple `closed-box' model.
1135:
1136: \acknowledgments
1137:
1138: Dimitrios A. Gouliermis\ acknowledges the support of the German Research
1139: Foundation (Deu\-tsche For\-schungs\-ge\-mein\-schaft - DFG) through the
1140: individual grant 1659/1-1. Boyke Rochau would like to thank Sascha P.
1141: Quanz and Alessandro Berton for their helpful suggestions. This paper is
1142: based on observations made with the NASA/ESA Hubble Space Telescope,
1143: obtained from the data archive at the Space Telescope Science Institute.
1144: STScI is operated by the Association of Universities for Research in
1145: Astronomy, Inc. under NASA contract NAS 5-26555.
1146:
1147: %\reference{}
1148: %%%%%%%%%%%%%%%%%%%%%%%%%%% BIBLIOGRAPHY %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1149: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1150: %\newpage
1151: \begin{references}
1152:
1153: \reference{} Alves, D.~R., \& Sarajedini, A.\ 1999, \apj, 511, 225
1154: \reference{} Baumgardt, H., \& Makino, J.\ 2003, \mnras, 340, 227
1155: \reference{} Bekki, K., Couch, W.~J., Beasley, M.~A., Forbes, D.~A., Chiba, M., \& Da Costa, G.~S.\ 2004, \apjl, 610, L93
1156: \reference{} Bertelli, G., Bressan, A., Chiosi, C., Fagotto, F., \& Nasi, E.\ 1994, \aaps, 106, 275
1157: \reference{} Bica, E.~L.~D., \& Schmitt, H.~R.\ 1995, \apjs, 101, 41
1158: \reference{} Binney, J., Merrifield, M., 1998, Galactic astronomy. Princeton University Press, Princeton, NJ
1159: \reference{} Binney, J., \& Tremaine, S., 1987, Galactic dynamics. Princeton University Press, Princeton, NJ
1160: \reference{} Crowl, H.~H., Sarajedini, A., Piatti, A.~E., Geisler, D., Bica, E., Clari{\'a}, J.~J., \& Santos, J.~F.~C., Jr.\ 2001, \aj, 122, 220
1161: \reference{} Da Costa, G.~S.\ 1991, IAU Symp.~148: The Magellanic Clouds, 148, 183
1162: \reference{} Da Costa, G.~S., \& Hatzidimitriou, D.\ 1998, \aj, 115, 1934
1163: \reference{} de Freitas Pacheco, J.~A., Barbuy, B., \& Idiart, T.\ 1998, \aap, 332, 19
1164: \reference{} de Grijs, R., Gilmore, G.~F., Johnson, R.~A., \& Mackey, A.~D.\ 2002, \mnras, 331, 245
1165: \reference{} Dolphin, A.~E.\ 2000, \pasp, 112, 1383
1166: \reference{} Dolphin, A.~E., Walker, A.~R., Hodge, P.~W., Mateo, M., Olszewski, E.~W., Schommer, R.~A., \& Suntzeff, N.~B.\ 2001, \apj, 562, 303
1167: \reference{} Elson, R.~A.~W., Fall, S.~M., \& Freeman, K.~C.\ 1987, \apj, 323, 54
1168: \reference{} Girardi, L., et al.\ 2002, \aap, 391, 195
1169: \reference{} Gouliermis, D., Keller, S.~C., Kontizas, M., Kontizas, E., \& Bellas-Velidis, I.\ 2004, \aap, 416, 137
1170: \reference{} Gouliermis, D.~A., Dolphin, A.~E., Brandner, W., \& Henning, T.\ 2006, \apjs, 166, 549 (Paper I)
1171: \reference{} Henize, K.~G.\ 1956, \apjs, 2, 315
1172: \reference{} Kayser, A., Grebel, E.~K., Harbeck, D.~R., Cole, A.~A., Koch, A., Gallagher, J.~S., \& Da Costa, G.~S.\ 2006, ArXiv Astrophysics e-prints, arXiv:astro-ph/0607047
1173: \reference{} Kerber, L.~O., \& Santiago, B.~X.\ 2006, \aap, 452, 155
1174: \reference{} Kroupa, P.\ 2002, Science, 295, 82
1175: \reference{} King, I.\ 1962, \aj, 67, 471
1176: \reference{} Lightman, A.~P., \& Shapiro, S.~L. \ 1978, Reviews of Morden Physics, 50, 437
1177: \reference{} McLaughlin, D.~E., \& van der Marel, R.~P.\ 2005, \apjs, 161, 304
1178: \reference{} Mackey, A.~D., \& Gilmore, G.~F.\ 2003a, \mnras, 338, 85
1179: \reference{} Mackey, A.~D., \& Gilmore, G.~F.\ 2003b, \mnras, 338, 120
1180: \reference{} Mackey, A.~D., \& Gilmore, G.~F.\ 2003c, \mnras, 340, 175
1181: \reference{} Mackey, A.~D., Payne, M.~J., \& Gilmore, G.~F.\ 2006, \mnras, 369, 921
1182: \reference{} Meylan, G., \& Heggie, D.~C.\ 1997, \aapr, 8, 1
1183: \reference{} Mighell, K.~J., Sarajedini, A., \& French, R.~S.\ 1998, \aj, 116, 2395
1184: \reference{} Naz{\'e}, Y., Hartwell, J.~M., Stevens, I.~R., Manfroid, J., Marchenko, S., Corcoran, M.~F., Moffat, A.~F.~J., \& Skalkowski, G.\ 2003, \apj, 586, 983
1185: \reference{} Naz{\'e}, Y., Manfroid, J., Stevens, I.~R., Corcoran, M.~F., \& Flores, A.\ 2004, \apj, 608, 208
1186: \reference{} Pagel, B.~E.~J., \& Tautvai\v sien\'e, G.\ 1998, \mnras, 299, 535
1187: \reference{} Peimbert, M., Peimbert, A., \& Ruiz, M.~T.\ 2000, \apj, 541, 688
1188: \reference{} Piatti, A.~E., Santos, J.~F.~C., Clari{\'a}, J.~J., Bica, E., Sarajedini, A., \& Geisler, D.\ 2001, \mnras, 325, 792
1189: \reference{} Piatti, A.~E., Sarajedini, A., Geisler, D., Seguel, J., \& Clark, D.\ 2005a, \mnras, 358, 1215
1190: \reference{} Piatti, A.~E., Santos, J.~F.~C., Jr., Clari{\'a}, J.~J., \, E., Ahumada, A.~V., \& Parisi, M.~C.\ 2005b, \aap, 440, 111
1191: \reference{} Pilyugin L. S., 1996, A\&A, 310, 751
1192: \reference{} Rafelski, M., \& Zaritsky, D.\ 2005, \aj, 129, 2701
1193: \reference{} Rich, R.~M., Shara, M., Fall, S.~M., \& Zurek, D.\ 2000, \aj, 119, 197
1194: \reference{} Rubio, M., Contursi, A., Lequeux, J., Probst, R., Barb{\'a}, R., Boulanger, F., Cesarsky, D., \& Maoli, R.\ 2000, \aap, 359, 1139
1195: \reference{} Russell S. C., \& Dopita M. A., 1992, ApJ, 384, 508
1196: \reference{} Sabbi, E., et al.\ 2007, \aj, 133, 44
1197: \reference{} Salpeter, E.~E.\ 1955, \apj, 121, 161
1198: \reference{} Scalo, J.~M.\ 1986, Fundamentals of Cosmic Physics, 11, 1
1199: \reference{} Spitzer, L., Jr.\ 1975, IAU Symp.~ 69: Dynamics of the Solar Systems, 69, 3
1200: \reference{} Spitzer, L.\ 1987, Princeton, NJ, Princeton University Press, p. 191
1201: \reference{} Tsujimoto T., Nomoto K., Yoshii Y., Hashimoto M., Yanagida S., Thielemann F.-K., 1995, MNRAS, 277, 945
1202: \reference{} van den Bergh, S.\ 1991, \apj, 369, 1
1203: \reference{} Vesperini, E., \& Heggie, D.~C.\ 1997, \mnras, 289, 898
1204: \reference{} Westerlund, B.~E.\ 1997, The Magellanic Clouds. Cambridge University Press, Cambridge
1205:
1206: \end{references}
1207: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1208:
1209: \end{document}
1210: