Astron. Astrophys. 320, 757-775 (1997)

1. Introduction

The advent of the charge-coupled devices technology and, very recently, the availability of the refurbished Hubble Space Telescope (HST ), coupled with the use of more and more sophisticated software for photometric analysis in crowded fields, have opened unique possibilities of using galactic globular clusters (GGCs) as templates for testing the stellar evolution theories (Renzini and Fusi Pecci 1988-RFP88). In particular, the combined use of both colour-magnitude diagrams (CMDs) and luminosity functions (LFs) derived from very accurate and complete photometric studies allows to reach levels of precision eventually suitable to check even the finer details predicted from updated theoretical models.

Since the basic aim is to secure stellar samples as populous and complete as possible in any radial region of a cluster to yield statistically significant information on all the stellar evolutionary phases (including the very short ones, yr) and on the cluster as a whole (f.i. taking into account also the cluster internal dynamics), proper ground-based observations in the outer areas must complement HST  observations exploiting the photometry in the very central regions.

Within this framework, we started a long-term project (ground-based + HST ) to make M 3 one of the first and best observed templates to verify in detail the model predictions (Buonanno et al.  1986, 1994, RFP88, Ferraro et al.  1993, Cacciari et al.  1993).

Why should M 3 justify such a big effort? There are many good reasons. Among others, for instance: a) since the early study by Sandage (1953), M 3 is one of the proto-type Pop II clusters of intermediate-poor metallicity; b) it is rich of RR Lyrae variables, and is usually adopted as Oosterhoff (1939) class I prototype; c) it contains the first detected blue straggler stars (BSS), now suspected to have a bimodal radial distribution (Ferraro et al.  1993); d) its CMD displays a significant population in every branch, with an Horizontal Branch (HB) spanning a very wide range in color (temperature), including extremely blue (hot) stars and Post-Asymptotic Giant Branch objects (P-AGB); e) it has been the target of recent spectroscopic studies (including at high resolution, see for references Carretta and Gratton 1996a), yielding a new insight on its average metallicity and metallicity dispersion.

The first step in our project has been a wide-field survey, based on the original Sandage's photographic plates, of regions from 2 up to 7 arcminutes from the cluster center, whose results have already been published (Buonanno et al.  1994-PH94). In that paper we published B,V magnitudes for more than 10,000 stars, reaching about two magnitudes fainter than the Main-Sequence turnoff (TO), with internal photometric errors mostly smaller than 0.05 mag. As a second step, in this paper we present the results of a new BVI CCD-photometry for almost 10,000 additional stars measured in the inner parts of the cluster. The third and final step to complete the survey will present the HST  data secured on the cluster core (proposal GO 5496, P.I. F. Fusi Pecci), whose reduction is currently in progress (Fusi Pecci et al.  1996). A further paper specifically devoted to the BVI CCD-photometry of 65 RR Lyrae variables is in preparation (Cacciari et al.  1996).

In Section 2, we present the description of data acquisition and analysis, including tests for completeness and comparisons with our previous photographic results (PH94) and with other data-sets. In Section 3, we describe our results, with special emphasis on the features of the main bright branches of the CMD and, in particular, we deal with mean ridge lines, the so-called RGB-Bump, star counts, population ratios, and an update of the overall blue stragglers population in M 3. Finally, Section 4 is devoted to conclusions and schematic summary.

© European Southern Observatory (ESO) 1997

Online publication: June 30, 1998
helpdesk.link@springer.de