Stefano Zibetti, Anna Gallazzi, Stephane Charlot, Daniele Pierini, Anna Pasquali
We present VLT-ISAAC NIR spectro-photometric observations of 16 post-starburst (PSB) galaxies aimed at constraining the debated influence of TP-AGB stars on the SED of galaxies with stellar ages between 0.5 and 2 Gyr, hence critical for high-redshift studies. PSB galaxies have negligible current star formation and a SED dominated by the stellar population formed in a recent (<2 Gyr) burst. By spectroscopically selecting PSB galaxies with mean luminosity-weighted ages between 0.5 and 1.5 Gyr and a broad range of metallicities, we explore the parameter space over which the relative energy output of TP-AGB stars peaks. A key feature of the present study is that we target galaxies at z~0.2, so that two main spectral features of TP-AGB stars (C-molecule band-head drops at 1.41 and 1.77mum, blended with strong telluric absorption features, hence hardly observable from the ground at z~0) move inside the H and K atmospheric windows and can be constrained for the first time to high accuracy. Our observations provide key constraints to stellar population synthesis models. Our main results are: i) the NIR regions around 1.41 and 1.77mum (rest-frame) are featureless for all galaxies in our sample at variance with the Maraston (2005) "TP-AGB heavy" models, which exhibit marked drops there; ii) no flux boosting is observed in the NIR: the optical-NIR SEDs of our PSB galaxies are generally consistent with Bruzual & Charlot (2003) simple stellar populations (SSP) of corresponding light-weighted ages and metallicities, but cannot be reproduced using Maraston (2005) SSPs. Possible systematic effects, including biases due to finite and different spectroscopic apertures, dust attenuation and, more importantly, the mixing of the pure post-burst stellar population with an old underlying component, are analysed and shown not to be able to reconcile observations and "TP-AGB heavy" models.
View original:
http://arxiv.org/abs/1205.4717
No comments:
Post a Comment