Astrocat Starcat Homepage


CATALOG OF PARAMETERS FOR MILKY WAY GLOBULAR CLUSTERS

Compiled by William E. Harris, McMaster University Last revision: June 1994

Source lists, calibration parameters, and conversion relations for the cluster data are described below.

CLUSTER IDENTIFICATIONS AND COORDINATES

The objects in this catalog adopted as Milky Way globular clusters essentially follow Djorgovski and Meylan, with the following exceptions:
  • AM-2 is excluded, since it is almost certainly an old open cluster (Gratton and Ortolani 1988, AASuppl. 74, 299).
  • Reticulum is excluded as more likely belonging to the LMC (e.g. Walker 1992, AJ 103, 1166).
  • Terzan 12 is restored to its original designation; it has mistakenly been labelled as Terzan 11 in several previous catalogs (including Kukarkin 1974, Harris and Racine 1979, Webbink 1985, and Djorgovski and Meylan 1993). To add to the potential confusion, the object number 11 in Terzan's lists is actually a rediscovery of Terzan 5. To avoid all such identification problems, it seems preferable to keep the original notation for these objects.
  • Djorgovski 3 has now been cross-identified as NGC 6540 (Djorgovski and Meylan 1994, AJ, in press). With the changes noted, the present catalog contains 141 objects adopted as certain or highly probable Milky Way globular clusters. It should be emphasized that some objects -- the ones with virtually no data entries in the catalog -- still have somewhat uncertain identities, and that the adoption of a final candidate list continues to be a matter of some arbitrary judgment for certain objects. The sources given above should be consulted for excellent discussions of these individually troublesome objects, as well as lists of other less likely candidates.

    The cluster coordinates are taken from Djorgovski and Meylan (1993).

    FOREGROUND REDDENING E(B-V)

    In addition to the three major sources listed above, measurements of E(B-V) from the individual color-magnitude studies (listed below) were employed whenever they appeared to be well calibrated [86 clusters]. These discussions, which employ a wide variety of semi-independent methods, now agree extremely well for all but a few cases. The final adopted reddenings are the straight averages of the given sources (up to 4 per cluster).

    DISTANCES: HORIZONTAL-BRANCH MAGNITUDE V_HB

    The primary distance indicator used here is the mean V magnitude of the horizontal branch (or RR Lyrae stars), V_HB. Wherever possible, the HB level is measured directly from a color-magnitude diagram (113 clusters); in a few cases where CMDs are not yet available, it can be evaluated from single-color photometry of the RR Lyrae variables (5 clusters), or more roughly predicted from the magnitude level of the brightest red giants (16 clusters). Whenever one of these latter two methods was used, it is noted in the list below.

    For many clusters, there are several CMD studies to be found in the literature, and in some cases the CMD study used here may not be the "best" one for the given cluster according to some other criterion. For example, it may not provide the best or most precise study of the main sequence, cluster age, metallicity, or some other feature. The sources listed below are simply those which, in the author's judgment, give the best available definition of the horizontal-branch level.

    The absolute calibration of V_HB adopted here uses a modest dependence of absolute V magnitude on metallicity,

                    M_V(HB) = 0.2 [Fe/H] + 1.00
    		

    which matches the body of recent evidence both from distance measurements of field RR Lyrae stars (e.g. Carney, Storm, and Jones 1992, ApJ, 386, 663; Skillen et al. 1993, MNRAS, 265, 30) and from theoretical stellar models of HB stars (e.g. Lee 1990, ApJ, 363, 159). The absolute uncertainty of the predicted M_V(HB) is +-0.1 mag at best, and may still be as large as +-0.2 mag in extreme cases. For a few clusters (mostly ones in the Galactic bulge region with very heavy reddening), no good [Fe/H] estimate is currently available; for these cases, a value M_V(HB) = 0.8 is simply assumed.

    INTEGRATED VISUAL MAGNITUDE V_t

    All data from the given sources are used, with the exception of the clusters specifically mentioned above (in these latter instances the V_t values are significantly discrepant compared with the other studies. The three major sources (Peterson and Reed, van den Bergh et al., Webbink) all use similar original databases (mostly concentric-aperture photometry of the clusters), but calculate total magnitudes independently by a variety of curve-of-growth methods.

    For several sparse or low-luminosity clusters (Pal 1, AM-1, E3, Pal 3, AM-4, Pal 14, Pal 15, Ter 12, Pal 12), the cluster luminosity function (from the color-magnitude study or other data in the literature) has been used by the present author to estimate V_t.

    The adopted integrated magnitudes are the straight averages of the data from all sources.

    INTEGRATED COLORS (U-B, B-V, V-R, V-I)

    The integrated colors U-B and B-V are on the standard Johnson system, and V-R, V-I on the Kron-Cousins system. The adopted values are the straight average of the two sources; see Peterson (1993) for a discussion of the (good) internal agreement of the databases and additional references.

    SPECIFIC FREQUENCY OF RR LYRAE STARS, S_RR

    S_RR is the number of RR Lyrae stars in the cluster, normalized to a total cluster luminosity M_Vt = -7.5; i.e.
                  S_RR = N_RR * 10**(0.4 (7.5 + M_Vt))
    
    The data source for N_RR is the comprehensive discussion of Suntzeff, Kinman, and Kraft 1991, ApJ, 367, 528. The resulting values for S_RR in the present catalog differ noticeably from theirs in some cases because of revisions in the cluster distance and (thus) luminosity.

    HORIZONTAL BRANCH MORPHOLOGY

    The HBR parameter defined by Lee (1990, ApJ 363, 159) is equal to (B-R)/(B+V+R), where in this case B denotes the number of horizontal-branch stars on the blue side of the RR Lyrae region; V the number of stars in the RR Lyrae region; and R the number of stars on the red side of the RR Lyrae region. The adopted HBR value is the direct average of the sources listed. The morphological type of the horizontal branch as originally defined by Dickens (1972, MNRAS 157, 281) is on an integer scale from 1 to 7, where 1 denotes a cluster with a completely blue HB (all stars on the blue side of the RR Lyrae region) and 7 denotes a completely red HB. From the individual color-magnitude studies listed above, the HB type has been evaluated anew by the present author for the purpose of this catalog. In addition, a new type "0" has also been introduced, to denote any cluster with an extremely blue extension of the HB. In a few cases, the distribution of stars along the HB is clearly bimodal, thus the cluster is assigned two simultaneous HB types.

    METALLICITIES: [Fe/H] VALUES

    The heavy-element abundance scale adopted here is basically the one established by Zinn and West (1984, ApJSuppl, 55, 45). In addition to the two major compilations of cluster metallicities listed below on the Zinn-West scale, data of two types are used: (1) [Fe/H] values from spectroscopy of individual cluster stars, from various more recent papers (listed below); and (2) metallicity estimates from the color-magnitude diagrams, in cases where the CMD is precise and well calibrated (references given in the V_HB section above). The final adopted [Fe/H] for each cluster is the straight average of the available measurements.

    RADIAL VELOCITY

    Most of the radial velocity data come from the four major sources listed below. However, numerous more recent sources are also available for smaller lists of objects; in many cases these are based on large samples of stars from CORAVEL or multi-object echelle spectra with very high precision (+-1 km/s or less) and almost totally supersede any previous data. The adopted v_r for each cluster is the average of the available measurements, each one weighted inversely as the published uncertainty. Unlike many globular cluster parameters, it is possible to quantify the uncertainty in v_r on an individual basis reasonably well, so the calculated uncertainty in the mean v_r is also given in the data table.

    Correction of the heliocentric velocity v_r to the Solar Local Standard of Rest is calculated by

     v_LSR = v_r + 10.4 cos b cos l + 14.8 cos b sin l + 7.3 sin b
    
    Additional Sources:

    STRUCTURAL AND DYNAMIC PARAMETERS

    The cluster core radii r_c, and the central concentration c = log(r_t/r_c), where r_t is the tidal radius, are taken from the comprehensive discussion of Trager et al. (1993). The half-mass radius r_h is taken from the direct average of Trager et al. and van den Bergh et al. There are 22 clusters for which Trager et al. quote values of c and r_c, but for which r_h is not given by either source. For these, estimates of r_h have been calculated from the approximate relation
                log (r_h/r_c) = 0.6 c - 0.4
    
    which is accurate to about 20%. These 22 objects are mostly sparse or low-luminosity clusters.

    The relaxation times in the core and at the half-mass radius, t(r_c) and t(r_h), are taken from Djorgovski (1993).

    The central surface brightness mu_V (magnitudes per square arcsecond) is the average of the values from Trager et al. (1993) and Webbink (1985). The central luminosity density, rho_0, in solar luminosities per cubic parsec, is calculated from the prescription given by Djorgovski (1993), using the previously determined values of the central surface brightness, total cluster luminosity, and core radius. These are different in some cases from the values listed by Djorgovski, because of revisions to the cluster distances.