Synthetic spectroscopic indices for identifying multiple stellar populations in globular clusters

Abstract

ABSTRACT We present an investigation of synthetic spectroscopic indices that can plausibly help in identifying the presence of multiple stellar populations in globular clusters. The study is based on a new grid of stellar model atmospheres and high-resolution (R = 500 000) synthetic spectra, that consider chemical partitions that have been singled out in Galactic globular clusters. The data base is composed of 3472 model atmospheres and theoretical spectra calculated with the collection of Fortran codes DFSYNTHE, ATLAS9 and SYNTHE, developed by Robert L. Kurucz. They cover a range of effective temperature from 4300 to 7000 K, surface gravity from 2.0 to 5.0 dex and four different chemical compositions. A set of 19 spectroscopic indices were calculated from a degraded version (R = 2500) of the theoretical spectra data set. The set includes five indices previously used in the context of globular clusters analyses and 14 indices that we have newly defined by maximizing the capability of differentiating the chemical compositions. We explored the effects of atmospheric parameters on the index values and identified the optimal spectral diagnostics that allow to trace the signatures of objects of different stellar populations, located in the main sequence, the red giant branch and the horizontal branch. We found a suitable set of indices, that mostly involve molecular bands (in particular NH, but also CH and CN), that are very promising for spectroscopically identifying multiple stellar populations in globular clusters.