A Survey of Open Clusters in the u'g'r'i'z' Filter System: III. Results for the Cluster NGC 188
ABSTRACT We continue our series of papers describing the results of a photometric survey of open star clusters, primarily in the southern hemisphere, taken in the u'g'r'i'z' filter system. The entire observed sample covered more than 100 clusters, but here we present data only on NGC 188, which is one of the oldest open clusters known in the Milky Way. We fit the Padova theoretical isochrones to our data. Assuming a solar metallicity for NGC 188, we find a distance of 1700+/-100 pc, an age of 7.5+/-0.7 Gyr, and a reddening E(B-V) of 0.025+/-0.005. This yields a distance modulus of 11.23+/-0.14.
arXiv:astro-ph/0611900v1 29 Nov 2006
Accepted by AJ
A Survey of Open Clusters in the u′g′r′i′z′Filter System: III.
Results for the Cluster NGC 188
Bartosz Fornal1,2,3, Douglas L. Tucker2,4, J. Allyn Smith4,5, Sahar S. Allam2,4,6, Cristin J.
Rider7, and Hwankyung Sung8
We continue our series of papers describing the results of a photometric survey
of open star clusters, primarily in the southern hemisphere, taken in the u′g′r′i′z′
filter system. The entire observed sample covered more than 100 clusters, but
here we present data only on NGC 188, which is one of the oldest open clusters
known in the Milky Way. We fit the Padova theoretical isochrones to our data.
Assuming a solar metallicity for NGC 188, we find a distance of 1700 ± 100 pc,
an age of 7.5±0.7 Gyr, and a reddening E(B −V ) of 0.025±0.005. This yields
a distance modulus of 11.23 ± 0.14.
Subject headings: Galaxy: open clusters and associations: individual: NGC 188,
Hertzsprung-Russell diagram, stars: abundances
1Institute of Physics, Jagiellonian University, ul. Reymonta 4, 30-059 Krak´ ow, Poland
2Fermi National Accelerator Laboratory, P.O. Box 500, Batavia, IL 60510
3Fermi National Accelerator Laboratory Internship for Undergraduate Physics Majors program
4Visiting Observer, United States Naval Observatory, Flagstaff Station, AZ
5Department of Physics & Astronomy, Austin Peay State University, P.O. Box 4608, Clarksville, TN
6University of Wyoming, Department of Physics & Astronomy, Laramie, WY 82071
7Department of Physics & Astronomy, Johns Hopkins University, 3400 N. Charles St., Baltimore, MD
8Department of Astronomy & Space Science, Sejong University, Seoul, 143-747, Korea
– 2 –
The study of open clusters is extremely important for improving our understanding of
stellar and galactic evolution. In particular, knowledge of accurate fundamental parameters
of star clusters (i.e., age, distance, reddening, metallicity) is essential for many astrophysical
calibrations. It also contributes to a better explanation of the present Galactic disk properties
and its past history. For example, precise measurements of age and distance of open clusters
from the Sun tell us about their spatial and age distributions, which give invaluable insight
into the disk structure. From these data it is possible to infer many characteristics of
the Milky Way, like the thin-disk scale height, horizontal scale length, or the displacement
of the Sun above the Galactic plane (see, Piskunov et al. 2006 and Bonatto et al. 2006).
Furthermore, data on reddening for open clusters show important features of the interstellar
extinction (Piskunov et al. 2006), which provides information on the Milky Way gas and dust
distribution. Correctly calculated open cluster parameters are therefore essential to constrain
Galactic theoretical models. Among all open clusters, the oldest ones are of special interest.
Their age can be established quite easily and they may be detected at large distances thanks
to their brightest members - strong-lined red giants (Friel 1995). Besides being fine tracers
of the present structure and chemistry of the Galactic disk, they enable us to look at disk
properties at epochs just after its formation, thus providing a reliable way to probe even the
early disk evolution. Accurately estimated old open cluster basic parameters are then very
useful for examining the disk-halo connection (Janes & Phelps 1994). As a result of being
such a powerful tool for testing theories of star formation and metal enrichment in the Milky
Way disk, open clusters, especially the old ones, have been the subject of intense studies for
the past several decades (besides the papers mentioned above, see, Anthony-Twarog et al.
1979; Geisler & Smith 1984; Anthony-Twarog et al. 1994; Bruntt et al. 1999; Mathieu 2000).
We undertook a survey of (mostly) southern hemisphere star clusters using the u′g′r′i′z′
filter system (Smith et al. 2002; Rider et al. 2004; Rodgers et al. 2006). The initial effort
in this survey delivered observations for approximately 105 open clusters and a few (less
than 10) “low-density” globular clusters. The original motivation of the project was to use
these clusters, which span a range of ages and metallicities, to “back calibrate” the Sloan
Digital Sky Survey (SDSS) (see, York et al. 2000 and Adelman-McCarthy et al. 2006 for a
description of the SDSS; Gunn et al. 2006 for a description of the SDSS 2.5 m telescope;
Gunn et al. 1998 for a description of the SDSS imaging camera; and Fukugita et al. 1996,
Hogg et al. 2001, Ivezic et al. 2004, and Tucker et al. 2006 for a description of the photo-
metric calibration of the SDSS.) We are now using these data to verify the recent age and
metallicity models presented in Girardi et al. (2004) and the prior work of Lenz et al. (1998)
and to verify and expand upon the u′g′r′i′z′to UBV RI transformations presented in the
SDSS standard star paper Smith et al. (2002). We have recently supplemented these data
– 3 –
with observations of northern hemisphere clusters using the United States Naval Observatory
1.0 m telescope.
In this, the third paper of our series, we present our results for the open cluster NGC 188.
We chose this particular cluster because it is one of the oldest open clusters known in our
galaxy. As described in Bonatto et al. (2005), it is located quite far from the galactic disk
and contains a couple of hundred member stars. Its field is not heavily contaminated by
background stars and it is also relatively free from dust (Bonatto et al. 2005). All this
makes NGC 188 a perfect target for testing stellar evolution models.
NGC 188 was discovered in 1831 by John Herschel. Since then, Ivan King noticed in
1948 that this cluster is very interesting since its angular diameter is large when compared
to the faint apparent magnitudes of its brightest stars (see, Sandage 1962a). In 1956 Sidney
van den Bergh obtained the first results suggesting that NGC 188 belongs to the group of
the oldest clusters (see, Sandage 1962a).
NGC 188 has been very carefully studied throughout the years, but the derivation of
its characteristic parameters has had an erratic history. Table 1 presents a summary of past
results on NGC 188. This table includes the reference paper (col. ), estimated distance
modulus, in the brackets the corresponding distance (corrected for the reddening E(B−V ))
(col. ), age (col ), E(B −V ) (col. ), metallicity [Fe/H] (col. ), the technique used
(col. ), and finally additional comments at the end.
One of the first studies of this cluster to determine age, distance and reddening is found
in Sandage (1962a) and Sandage (1962b). In these two papers, Sandage estimated the age
of this cluster to be 14−16 Gyr (based on Hoyle 1959 stellar models), the distance modulus
m − M to be 10.95, and the reddening E(B − V ) to be between 0.03 and 0.07, all these
values depending on the chemical composition.
Revised values of NGC 188 parameters based on UBV photometry are presented in
Eggen & Sandage (1969), who give an age of 10 Gyr, a distance modulus of 10.85, a reddening
E(B − V ) of 0.09, and a metallicity [Fe/H] of 0.07. During the following two decades
estimates of the age varied from 3.6 Gyr (Torres-Peimbert 1971) and 4.3 Gyr (Twarog 1978)
up to 10 Gyr (VandenBerg 1985, Adler & Rood 1985). Estimates of the metallicity ranged
from [Fe/H] = −0.6 (Jennens & Helfer 1975) to as much as [Fe/H] = 0.6 (Spinrad et al.
1970). The estimates for the distance modulus fell within the range of 10.8 (Patenaude 1978)
to 12.0 (Jennens & Helfer 1975), while estimates for the reddening E(B − V ) varied from
0.04 (Jennens & Helfer 1975) to 0.15 (Spinrad et al. 1970).
The analysis of color-magnitude diagrams performed by Twarog & Anthony-Twarog
(1989) yielded an age between 6 and 7 Gyr, a distance modulus of 11.50 and a high E(B−V )
– 4 –
value of 0.12. Besides the Hobbs et al. (1990) value of E(B − V ) = 0.10 no other esti-
mation of the reddening was made until Carraro et al. (1994) paper which yielded values
0.03 and 0.04, depending on the adopted stellar evolution model. The value of age varied
from 6 Gyr (Paez et al. 1990, Dinescu et al. 1995) to 7.7 Gyr (Hobbs et al. 1990). Estima-
tions of distance modulus ranged from just a little over 11.0 (Branly et al. 1996) up to 11.5
(Dinescu et al. 1995). The estimated metallicity was always close to solar.
A more recent, high-precision UBV RI CCD photometry study by Sarajedini et al.
(1999) provided a color-magnitude diagram which extends almost from the red giant branch
tip to approximately 5 mag below the main sequence turn-off. The final conclusion was that
there is a considerable offset between the photometric zero point of these results and those
from Eggen & Sandage (1969), whose photometric scale was used in all previous photometric
studies of NGC 188. Reddening E(B − V ) was found to be equal to 0.09 ± 0.02 and the
distance modulus to be 11.44±0.08. The metallicity was assumed to be solar, based on the
result [Fe/H] = −0.04 ± 0.05 from von Hippel & Sarajedini (1998). The fitted isochrones
yielded an age of 7.0 ± 0.5 Gyr. Moreover, the data indicate that there exists a mass segre-
gation with the most massive stars (M/M⊙> 1.1) more centrally concentrated comparing
to the least massive ones (0.8 > M/M⊙> 0.65).
A metallicity for NGC 188 close to solar was verified in Friel et al. (2002), Randich et al.
(2003), and Worthey & Jowett (2003). The values of other parameters were revised in
three of the most recent papers — VandenBerg & Stetson (2004), Bonatto et al. (2005),
and Haroon et al. (2004) — by fitting theoretical isochrones to the color-magnitude dia-
gram. In the first one, after adopting a reddening E(B − V ) of 0.087 from Schlegel et al.
(1998) dust maps, VandenBerg & Stetson (2004) determined an age of 6.8 Gyr and a dis-
tance modulus of 11.40 (distance 1685 pc). In the second of these papers, Bonatto et al.
(2005) found a slightly higher age of 7.1 Gyr and a smaller distance modulus of 11.10; their
best-fit isochrone indicates E(B − V ) = 0.00 (which gives a distance of 1660 pc). Finally,
both of these age values fall within the range stated in the third paper, Haroon et al. (2004).
The results therein indicate an age between 6 and 8 Gyr, more likely close to 8 Gyr. The
fitted isochrone yields a distance modulus of 11.26 ± 0.05. An abnormally high value of
reddening, E(B − V ) = 0.17, lowers the resulting distance down to only 1415 ± 35 pc.
The most recent star catalog for NGC 188 is presented in Stetson et al. (2004). It
contains detailed information on more than 9000 stars in the field of the cluster based on all
available studies. In addition, half of those stars have revised photometry.
Ernst Paunzen and Jean-Claude Mermilliod’s webda online open cluster database,1
– 5 –
which provides a compilation of data from several sources, currently (August 2006) lists
the age of NGC 188 as 4.3 Gyr with a distance of 2047 pc, a distance modulus of 11.81, a
reddening of E(B − V ) = 0.082, and a metallicity of [Fe/H] = −0.02.
In the following sections of this paper we present details of the instrumentation and
observations (§2), data reduction and analysis techniques (§3), isochrones (§4), results (§5),
and a summary (§6).
2. Instrumentation and Observations
The five filters of the u′g′r′i′z′system have effective wavelengths of 3540˚ A, 4750˚ A,
6222˚ A, 7632˚ A, and 9049˚ A, respectively, at 1.2 airmasses.2They cover the entire wavelength
range of the combined atmosphere and CCD response. Their construction is described in
Fukugita et al. (1996). The most important characteristics of the u′g′r′i′z′filters and the
u′g′r′i′z′magnitude system are outlined in Rider et al. (2004). For a more detailed explana-
tion they refer to Oke & Gunn (1983), Fukugita et al. (1996), and Smith et al. (2002). The
u′g′r′i′z′standard star network consists of 158 stars distributed primarily along the celestial
equator and the northern celestial hemisphere Smith et al. (2002). Efforts are in place and
nearing completion for extending this network both fainter and redder (Smith et al. 2007)
and into the southern hemisphere (Smith et al. 2006).
2.2. Telescope and Observations
For this paper, we are using data from the USNO 1.0 m telescope. The observations
for NGC 188 were obtained on 2002 November 5 (UT) as part of a four-night observing run.
An overview of the observing circumstances is given in Table 2.
All of the observations were direct exposures with a thinned, UV-AR coated, Tektronix
TK1024 CCD operating at a gain of 7.43 ± 0.41 e−ADU−1with a read noise of 6.0 e−. This
CCD is similar to the CCDs used in the SDSS 2.5 m telescope’s imaging camera and the
CCD used by the 0.5 m SDSS Photometric Telescope at Apache Point Observatory. This
is the detector that defines the u′g′r′i′z′standard star network. The camera scale of 0.”68
2Note that the g′filter has been determined to have an effective wavelength 20˚ A bluer than that originally
quoted by Fukugita et al. (1996).