FIGS: Spectral fitting constraints on the star formation history of massive galaxies since Cosmic Noon

TL;DR

This study uses deep HST slitless spectra to analyze the stellar populations of 52 massive galaxies between redshifts 0.5 and 2, revealing insights into their star formation histories and age distributions.

Contribution

It introduces a spectral fitting method tailored to redshift and spectral resolution, applied to a diverse galaxy sample to constrain their star formation histories.

Findings

Older stellar populations at higher masses.

Weak anti-correlation between age spread and stellar mass.

Redshift dependence rejects monolithic collapse hypothesis.

Abstract

We constrain the stellar population properties of a sample of 52 massive galaxies, with stellar mass log Ms>10.5, over the redshift range 0.5<z<2 by use of observer-frame optical and near-infrared slitless spectra from HST's ACS and WFC3 grisms. The deep exposures (~100 ks) allow us to target individual spectra of massive galaxies to F160W=22.5AB. Our spectral fitting approach uses a set of six base models adapted to the redshift and spectral resolution of each observation, and fits the weights of the base models, including potential dust attenuation, via an MCMC method. Our sample comprises a mixed distribution of quiescent (19) and star-forming galaxies (33). We quantify the width of the age distribution (Dt) that is found to dominate the variance of the retrieved parameters according to Principal Component Analysis. The population parameters follow the expected trend towards older ages with increasing mass, and Dt appears to weakly anti-correlate with stellar mass, suggesting a more efficient star formation at the massive end. As expected, the redshift dependence of the relative stellar age (measured in units of the age of the Universe at the source) in the quiescent sample rejects the hypothesis of a single burst (aka monolithic collapse). Radial colour gradients within each galaxy are also explored, finding a wider scatter in the star-forming subsample, but no conclusive trend with respect to the population parameters.