Astron. Astrophys. 355, 966-978 (2000)

2. The observational sample

Thirty-nine clusters have been observed with the ESO/Dutch 0.9m telescope at La Silla, and 16 at the RGO/JKT 1m telescope in la Palma. This database comprises of the GGC whose distance modulus is . The zero-point uncertainties of our calibrations are  mag for each band. Three clusters were observed both with the southern and the northern telescopes, thus providing a consistency check of the calibrations: no systematic differences were found, at the level of accuracy of the zero-points. A detailed description of the observations and reduction procedures will be given in forthcoming papers (Rosenberg et al. 1999b, 1999c) presenting the single clusters.

A subsample of this database was used for the present investigation. We retained those clusters whose CMD satisfied a few criteria: (a) the HB level could be well determined; (b) the RGB was not heavily contaminated by foreground/background contamination; and (c) the RGB was well defined up to the tip. This subsample largely overlaps that used for the age investigation, but a few clusters whose TO position could not be measured, are nevertheless useful for the metallicity indices definition. Conversely, in a few cases the lower RGB could be used for the color measurements, while the upper branch was too scarcely defined for a reliable definition of the fiducial line. Two of the CMDs that were used are shown in Figs. 1 (NGC 1851) and 2 (NGC 104), and they illustrate the good quality of the data.

Fig. 1. Graphical representation of the metallicity indices (part 1) that were measured on the selected clusters. (Left) The observed CMD of the intermediate-metallicity cluster NGC1851 and its fiducial RGB (solid line). The fiducial locus was obtained by fitting Eq. (1) to the data. The two crosses mark the color of the RGB at the level of the HB, and its color 2 mags brighter than the HB. The slope of the line connecting the two points is the S index. (Right) On the color de-reddened CMD other four indices are marked. From fainter to brighter magnitudes, the RGB color at the level of the HB, and the V magnitude difference between this point and those at , 1.2 and 1.4. . The dashed line represents the adopted HB level,

Fig. 2. Graphical representation of the metallicity indices (part 2) that were measured on the selected clusters. In this figure, the absolute CMD of the metal rich cluster NGC 104 is plotted in the plane, adopting an apparent distance modulus and a reddening (see text for the discussion). This plot shows the ability of the analytic function to reproduce even the more extended RGBs. The two crosses mark the color of the RGB at and .

The dataset of 31 clusters used in this paper is listed in Table 1. From left to right, the columns contain the NGC number, the reddening both in and , the metallicity according to three different scales, and the apparent magnitude of the horizontal branch (HB). The values were taken from the Harris (1996) on-line table ¹. The reddenings were obtained by assuming that (Dean et al. 1978). The values of the metallicity were taken from RHS97: they represent the equivalent widths of the CaII infrared triplet, calibrated either onto the Zinn & West (1984) scale (ZW column) or the Carretta & Gratton (1997) scale (RHS97 column). Moreover, the original Carretta & Gratton metallicities (CG column) are also given for the clusters comprised in their sample.

Table 1. The input parameters for the observational sample

The HB level was found in different ways for clusters of different metallicity. For the the metal rich and metal intermediate clusters, a magnitude distribution of the HB stars was obtained, and the mode of the distribution was taken. Where the HB was too scarcely populated, a horizontal line was fitted through the data. The blue tail of the metal poorest clusters does not reach the horizontal part of the branch: in that case, a fiducial HB was fitted to the tail, and the magnitude of the horizontal part was taken as the reference level. The fiducial branch was defined by taking a cluster having a bimodal HB color distribution (NGC 1851, cf. Fig. 1) and then extending its HB both to the red and to the blue by "appending" clusters being more and more metal rich and metal poor, respectively. The details of this procedure, as well as the errors associated to the in Table 1, are discussed in RSPA99. For NGC 1851, was adopted (dashed line in Fig. 1), and this value is just 0.02 mag brighter than the value found by Walker (1992) and Saviane et al. (1998).

Based on this observational sample, a set of metallicity indices were measured on the RGBs of the clusters. In the next section, the indices are defined and the measurement procedures are described. Consistency checks are also performed.

© European Southern Observatory (ESO) 2000

Online publication: April 3, 2000
helpdesk.link@springer.de