Integrated K-band spectra of old and intermediate-age globular clusters in the Large Magellanic Cloud

TL;DR

This study provides the first empirical near-IR spectral library of LMC globular clusters, revealing discrepancies with models related to carbon-rich TP-AGB stars and offering calibration data for stellar population models.

Contribution

It introduces a new empirical spectral library of LMC globular clusters for testing and calibrating stellar population models in the near-IR range.

Findings

Na I index decreases with age and metallicity.

Dco index is affected by carbon-rich TP-AGB stars in ~1 Gyr clusters.

Discrepancies with models due to different CO absorption in LMC carbon stars.

Abstract

Current stellar population models have arguably the largest uncertainties in the near-IR wavelength range, partly due to a lack of large and well calibrated empirical spectral libraries. In this paper we present a project, which aim it is to provide the first library of luminosity weighted integrated near-IR spectra of globular clusters to be used to test the current stellar population models and serve as calibrators for the future ones. Our pilot study presents spatially integrated K-band spectra of three old (>10 Gyr) and metal poor ([Fe/H]~-1.4), and three intermediate age (1-2 Gyr) and more metal rich ([Fe/H]~-0.4) globular clusters in the LMC. We measured the line strengths of the Na I, Ca I and 12CO(2-0) absorption features. The Na I index decreases with the increasing age and decreasing metallicity of the clusters. The Dco index, used to measure the 12CO(2-0) line strength, is significantly reduced by the presence of carbon-rich TP-AGB stars in the globular clusters with age ~1 Gyr. This is in contradiction with the predictions of the stellar population models of Maraston (2005). We find that this disagreement is due to the different CO absorption strength of carbon-rich Milky Way TP-AGB stars used in the models and the LMC carbon stars in our sample. For globular clusters with age >2 Gyr we find Dco index measurements consistent with the model predictions.