2. Observations and reduction
The observations in V and I were carried out at La Silla/Chile between July 16th and 19th 1993. We used the 2.2m with CCD ESO #19, which covers with pix an area of on the sky. The seeing was . In addition to the ground-based data, we used data of the Hubble-Space-Telescope (HST) for NGC 5927. These data have already been published by Fullton et al. (1996).
To reduce the data, we used the DAOPHOT-package (Stetson 1987, 1992), together with the ESO-MIDAS-system (version 1996). The calibrating equations (1), determined via Landolt standard stars (Landolt 1992), are
Together with the error of the photometry and of the PSF-aperture-shift, we get an absolute error for a single measurement of mag. For a more extensive treatment of the data and their reduction, see Heitsch & Richtler (1999). To calibrate the HST-data, we used the relations and coefficients as described by Holtzman et al. (1995). In agreement with Guarnieri et al. (1998), we detected a systematic shift between calibrated ground-based and HST-magnitudes of about 0.2 mag with the HST-magnitudes being fainter. This difference might be due to crowding influences on the calibration stars in the ESO-frame, as explained by Guarnieri et al.
The unselected CMD for NGC 5927 is shown in Fig. 6. The cluster's HB and RGB are clearly distinguishable, with the HB overlapping the RGB, as well as the stars of some field population to the blue of the cluster's structures covering the TOP-region. Some 0.5 mag below the HB, the RGB-bump is discernible. The elongation of the HB and the broadening of the RGB are due to differential reddening, as we now argue. As the HB of metal-rich GCs generally is rather clumped and the HB-stars all have the same luminosity, differential reddening should cause an elongation of the HB parallel to the reddening vector.
Fig. 2 shows the coordinates of the radially selected cluster stars with special markings for the HB-stars as given by Fig. 1. If differential reddening is indeed responsible for the observed elongation, we do not expect to find any red faint stars in areas where blue bright stars are to be found, unless the reddening is very (!) patchy. In the case of NGC 5927, we note (Fig. 2) that the blue bright stars are located in an area west of the cluster's center, and thus differential reddening is indeed responsible for the elongated HB structure.
In Fig. 20 we present the calibrated HST-CMDs of NGC 5927. They all show a slightly broadened lower RGB as well as a slightly tilted HB. The TOP is well resolved. However, the CMDs do not extend to the bright stars of the AGB/RGB due to pixel overflow on the exposures. As mentioned above, the HST-CMDs are shifted with respect to the ESO-CMDs to fainter magnitudes. Richtler et al. (1998) argue that the ground-based calibration is not erroneous. Thus, we will use the ground-based calibrationed data for the further analysis. For a detailed discussion see Heitsch & Richtler (1999).
The unselected CMD (Fig. 8) shows beside the cluster a strong contribution from the field population. The field main sequence is striking. The cluster RGB is broadened, but since the clumpy HB indicates only small differential reddening, the RGB width is probably to a large part due to the field contamination. Determining a correlation between HB-stars and coordinates as in Fig. 2 led to no convincing results, because the field does not contain enough stars.
NGC 6342 (Fig. 10) shows a sparsely populated AGB/RGB due to the small size of the cluster. The TOP-region and upper MS are reached. Fig. 3 gives the location of HB-stars as in Fig. 2. Blue, bright HB-stars are to be found in an area to the south of the cluster's center.
NGC 6441 (Fig. 12) is located behind a dense field population, the stars of which can be found between mag. This population covers the lower part of the RGB of NGC 6441 as well. Its TOP is not reached. We mention some special features. First, we find some stars between mag and above the HB of NGC 6441. These could be HB-stars of a population which is similar to NGC 6441 and is located between the cluster and ourselves, as the stars are shifted in V only. In this case, we would have to assume that the absolute reddening is caused by some cloud between this population and the observer. Otherwise it would have to be shifted in as well. Second, we find some stars to the blue of the clumpy, tilted HB of NGC 6441. These stars seemingly belong to the cluster, as they are still visible when selecting for small radii. Probably they belong to the blue HB of NGC 6441, which has been discovered by Rich et al. (1997). The difference in star density (Fig. 4) is due to the fact that the calibration exposures were shifted by around 360 pix to the south. Taking this into account, we find that the blue bright stars are mostly found in the western two thirds of the cluster.
2.1. NGC 6760
Fig. 14 not only shows the already discussed structures such as HB, RGB and field population, but it shows the RGB-bump below the HB as well. The RGB is rather broadened. The HB and RGB-bumps are elongated and tilted with the same slope. The HB-stars, marked according to colour and brightness, are found in Fig. 5.
2.2. NGC 6528 and NGC 6553
The CMDs for NGC 6528 and 6553 (Fig. 16, 18) have already been published (Richtler et al. 1998, Sagar et al. 1998). As described in their papers, NGC 6528 not only shows a broadened RGB and a tilted and elongated HB, but it shows also some background population below the AGB/RGB. The field population covers the TOP-region of the cluster. Moreover, the RGB-bump of NGC 6528 is clearly visible some 0.5 mag below the HB. The CMD of NGC 6553 shows the same characteristics as NGC 6528, however they are even more distinct. Here we clearly see the background population with its RGB and AGB/RGB strongly differentially reddened.
© European Southern Observatory (ESO) 1999
Online publication: June 30, 1999