Ultraviolet-to-far-infra-red self-consistent analysis of the stellar populations of massive starburst galaxies at intermediate redshifts

Abstract

We study in detail the properties of the stellar populations of 111 massive ($\log (M_{\star }/{\mathrm{M}}_{\odot })\ge 10)$ dusty (FIR-selected) starburst ($SFR/SF{R}_{\mathrm{MS}}>2$) galaxies at $0.7<z<1.2$. For that purpose, we use self-consistent methods that analyse the UV-to-FIR broadband observations in terms of the stellar light and dust re-emission with energy-balance techniques. We find that the emission of our starburst galaxies can be interpreted as a recent star formation episode superimposed on a more evolved stellar population. On average, the burst age is $\sim 80$ Myr and its attenuation $\sim 2.4$ mag. Assuming our starburst galaxies at half their lifetimes, we infer a duration of the starburst phase of $\sim 160$ Myr. The median stellar mass and SFR are $\log (M_{\star }/{\mathrm{M}}_{\odot })\sim 10.6$ and $\sim 220{\mathrm{M}}_{\odot }$yr${}^{-1}$. Assuming this SFR and the inferred duration of the starburst phase, the stellar mass added during this phase corresponds to $\sim 40$ per cent the median stellar mass of our sample. The young-population age determines the position of our galaxies in the $M_{\star }-SFR$ plane. Galaxies located at the largest distances of the MS present shorter young-population ages. The properties of the underlying stellar population cannot be constrained accurately with our broadband data. We also discuss the impact of including the FIR data and energy-balance techniques in the analysis of the properties of the stellar populations in starburst galaxies.

24 pages, 11 figures, MNRAS accepted