A near-infrared study of AGB and red giant stars in the Leo I dSph galaxy

TL;DR

This study uses near-infrared imaging to analyze evolved stellar populations in the Leo I dwarf galaxy, deriving metallicity, distinguishing AGB star types, and testing stellar evolution models.

Contribution

It provides new near-infrared data and analysis of AGB stars in Leo I, offering constraints on low-metallicity stellar evolution models and revealing population gradients.

Findings

Derived metallicity distribution consistent with spectroscopy.

Identified a radial gradient in intermediate-age stellar populations.

Compared observed AGB stars with models, revealing discrepancies in C star predictions.

Abstract

A near-infrared imaging study of the evolved stellar populations in the dwarf spheroidal galaxy Leo I is presented. Based on JHK observations obtained with the WFCAM wide-field array at the UKIRT telescope, we build a near-infrared photometric catalogue of red giant branch (RGB) and asymptotic giant branch (AGB) stars in Leo I over a 13.5 arcmin square area. The V-K colours of RGB stars, obtained by combining the new data with existing optical observations, allow us to derive a distribution of global metallicity [M/H] with average [M/H] = -1.51 (uncorrected) or [M/H] = -1.24 +/- 0.05 (int) +/- 0.15 (syst) after correction for the mean age of Leo I stars. This is consistent with the results from spectroscopy once stellar ages are taken into account. Using a near-infrared two-colour diagram, we discriminate between carbon- and oxygen-rich AGB stars and obtain a clean separation from Milky Way foreground stars. We reveal a concentration of C-type AGB stars relative to the red giant stars in the inner region of the galaxy, which implies a radial gradient in the intermediate-age (1-3 Gyr) stellar populations. The numbers and luminosities of the observed carbon- and oxygen-rich AGB stars are compared with those predicted by evolutionary models including the thermally-pulsing AGB phase, to provide new constraints to the models for low-metallicity stars. We find an excess in the predicted number of C stars fainter than the RGB tip, associated to a paucity of brighter ones. The number of O-rich AGB stars is roughly consistent with the models, yet their predicted luminosity function is extended to brighter luminosity. It appears likely that the adopted evolutionary models overestimate the C star lifetime and underestimate their K-band luminosity.