A VST and VISTA study of globular clusters in NGC 253

Context. Globular clusters (GCs) are key to our understanding of the Universe, as laboratories of stellar evolution, fossil tracers of the past formation epoch of the host galaxy, and effective distance indicators from local to cosmological scales. Aims. We analyze the properties of the sources in the NGC253 with the aim of defining an up to date catalog of GC candidates in the galaxy. Given the distance of the galaxy, GCs in NGC253 are ideal targets for resolved color-magnitude diagram studies of extragalactic GCs with next-generation diffraction limited ground-based telescopes. Methods. Our analysis is based on the science verification data of two ESO survey telescopes, VST and VISTA. Using ugri photometry from VST and JK(s) from VISTA, GC candidates were selected using as reference the morpho-photometric and color properties of spectroscopically confirmed GCs available in the literature. The strength of the results was verified against available archival HST/ACS data from the GHOSTS survey: all but two of the selected GC candidates appear as star clusters in HST footprints. Results. The adopted GC selection leads to the definition of a sample of similar to 50 GC candidates. At visual inspection, we find that 82 objects match all the requirements for selecting GC candidates and 155 are flagged as uncertain GC candidate; however, 110 are unlikely GCs, which are most likely background galaxies. Furthermore, our analysis shows that four of the previously spectroscopically confirmed GCs, i.e., similar to 20% of the total spectroscopic sample, are more likely either background galaxies or high-velocity Milky Way stars. The radial density profile of the selected best candidates shows the typically observed r(1/4)-law radial profile. The analysis of the color distributions reveals only marginal evidence of the presence of color bimodality, which is normally observed in galaxies of similar luminosity. The GC luminosity function does not show the typical symmetry, mainly because of the lack of bright GCs. Part of the bright GCs missing might be at very large galactocentric distances or along the line of sight of the galaxy dusty disk. As an alternative possibility, we speculate that a fraction of low luminosity GC candidates might instead be metal-rich, intermediate age clusters, but fall in a similar color interval of old, metal-poor GCs. Conclusions. Defining a contaminant-free sample of GCs in extragalactic systems is not a straight forward exercise. Using optical and near-IR photometry we purged the list of GCs with spectroscopic membership and photometric GC candidates in NGC 253. Our results show that the use of either spectroscopic or photometric data only does not generally ensure a contaminant-free sample and a combination of both spectroscopy and photometry is preferred.