4 Color-magnitude diagram

The color-magnitude diagrams of V versus B-V, V-R and V-Iare presented in Figs. 6-8, respectively, with different symbols for different membership probabilities as used in Fig. 5.

Figure 6: Diagram of V vs. B-V

Figure 7: Diagram of V vs. V-R

Figure 8: Diagram of V vs. V-I, *: stars located on HB or GB and constitute a test sample

The characteristic behaviour of these diagrams is, on the whole, similar to the CMD of V versus B-V from Aurière & Lauzeral (1991), however with a large dispersion for the cluster members, especially there are quite a few stars with large membership probabilities located above the horizontal branch. From our CMDs of V versus V-R and V-I, it seems that there exists a second horizontal branch (somewhat similar to the second main sequence in CMD of some open clusters). It might be caused by double stars. On the other hand in our sample, we have not identified the RR Lyrae stars which would drift off the horizontal branch. Moreover, crowding problems near the cluster center caused the reduced accuracy of the photometric data, and this could be another reason for the large dispersion in the member star sequence. In our sample 54 stars, which appeared only on one of the first-epoch plates (CL 58004) and not in the catalogue of Geffert et al. (1997), have relatively large uncertainties in the membership probabilities since their proper motion estimates can only be regarded as preliminary. In addition, some of our stars with large membership probabilities have large internal proper motion errors (which can be seen in the vector point diagram of proper motions). In this case, it is likely that one or two non-members were estimated as member stars, which could also lead to an increase of the dispersion in the CMDs. It is interesting to note that the star sorted out by Aurière & Lauzeral (1991), being brighter than the red giants by about 1 mag and redder by 0.4 mag is, according to our result, a cluster member with a membership probability of 0.97. It is located on the extension of the red giant branch in all three CMDs. Based on this, it is very probably an AGB star which is entering or has entered already the thermal pulsation stage. Since the cluster radial velocity is large, and the star not exceedingly faint, a future radial velocity measurement may check its membership.

  

Table 9: Absolute proper motions and BVRI photometry of 115 stars in the region of globular cluster NGC 4147

  

Table 9: continued

  

Table 9: continued

  

Table 9: continued

  

Table 9: continued