3. Parallaxes and proper motions for 9 open clusters
The catalogue of open-cluster parameters by Lyngå (1987) contains 12 clusters for which the distance given is less than 250 pc. At this distance, a cluster with stellar density like the Pleiades will still cover an area of about 4 degrees diameter on the sky (some 12 square degrees), making it likely to find at least of the order of 5 to 10 members amongst the Hipparcos stars. Given an average parallax accuracy of 1.2 mas, this would provide an estimated formal error for the mean cluster parallax of 0.3 to 0.4 mas, on a parallax of 4 mas or more. Thus, the 250 pc limit corresponds roughly with the 10 percent accuracy level for the expected cluster parallax, or, when expressed in distance modulus, a formal error of at most 0.2 magnitudes. The estimated accuracies for the distances in the Lyngå catalogue are about 10 per cent, originating from pre-Hipparcos Hyades distance calibrations and subsequent main sequence fits.
Of the 12 selected clusters, the Hyades has been extensively studied by Perryman et al. (1998) and Madsen (1999). Due to its extent on the sky, its study requires special methods which are not provided for in the reduction software used for the more distant clusters. A re-examination of the Hyades is therefore not included. Of the remaining 11 clusters, two were clearly shown to be spurious: Upgren 1 (C1232+365) and Collinder 399 (C1923+200). The cluster parameters and selection criteria for the remaining 9 clusters are presented in Table 1.
In obtaining selections of cluster members it was noted that for most clusters an additional proper motion dispersion had to be included (see column 9 in Table 1), which was in most cases larger than could be expected from internal dynamics. However, in the combined solution there was in all but two cases (Pleiades and Praesepe) no trace left from this internal dispersion: the unit weight standard error of each solution was very close to the expected value of 1.00. The additional dispersion was therefore more likely due to correlations within the astrometric parameter solutions of the individual stars. In the Pleiades an internal proper motion dispersion at a level of 0.8 to 1.0 mas per year has been observed from ground-based differential proper motions (van Leeuwen, 1994), while in Praesepe the selection of members could be affected by a halo of escaped members as has been observed for the Hyades by Perryman et al. (1998). The fact that in most solutions the standard error was below one is an indication that on average the correlations of the abscissa residuals may have been slightly overestimated, which would mean that the formal errors on the mean astrometric cluster parameters are slightly overestimated. It should also be noted that, though included here, Per clearly showed not to be dynamically bound as a cluster, and therefore should not be considered to be equally homogeneous in age and chemical composition as the real clusters included.
A comparison between the distance estimates as given by Lyngå (1987) and as obtained from the Hipparcos data shows an overall good agreement, as can be seen in Fig. 1, where it should be noted that more recent estimates of the Pleiades distance have in general put the cluster at 130 rather than 125 pc (Pinsonneault et al., 1998 and references therein).
© European Southern Observatory (ESO) 1999
Online publication: December 16, 1998