Proper motions in the VVV Survey: Results for more than 15 million stars across NGC 6544

Abstract

Context.In the last six years, the VISTA Variable in the Vía Láctea (VVV) survey mapped 562 sq. deg. across the bulge and southerndisk of the Galaxy. However, a detailed study of these regions, which includes∼36 globular clusters (GCs) and thousands of openclusters is by no means an easy challenge. High differential reddening and severe crowding along the line of sight makes highlyhamper to reliably distinguish stars belonging to different populations and/or systems.Aims.The aim of this study is to separate stars that likely belong to the Galactic GC NGC 6544 from its surrounding field by meansof proper motion (PM) techniques.Methods.This work was based upon a new astrometric reduction method optimized for images of the VVV survey.Results.PSF-fitting photometry over the six years baseline of the survey allowed us to obtain a mean precision of∼0.51 mas yr−1, ineach PM coordinate, for stars withKs<15 mag. In the area studied here, cluster stars separate very well from field stars, down to themain sequence turnoffand below, allowing us to derive for the first time the absolute PM of NGC 6544. Isochrone fitting on the cleanand differential reddening corrected cluster color magnitude diagram yields an age of∼11−13 Gyr, and metallicity [Fe/H]=−1.5 dex,in agreement with previous studies restricted to the cluster core. We were able to derive the cluster orbit assuming an axisymmetricmodel of the Galaxy and conclude that NGC 6544 is likely a halo GC. We have not detected tidal tail signatures associated to thecluster, but a remarkable elongation in the galactic center direction has been found. The precision achieved in the PM determinationalso allows us to separate bulge stars from foreground disk stars, enabling the kinematical selection of bona fide bulge stars across thewhole survey area.Conclusions.Kinematical techniques are a fundamental step toward disentangling different stellar populations that overlap in astudied field. Our results show that VVV data is perfectly suitable for this kind of analysis.