http://arxiv.org/abs/1412.2760
(abridged) Significant evidence has been gathered suggesting the existence of a main sequence (MS) of star-forming galaxies that relates their star formation rate and their stellar mass: $SFR \propto M_*^{\alpha}$. Several ideas have been suggested to explain fundamental properties of the MS, such as its slope, its dispersion, and its evolution with redshift. However, no consensus has been reached regarding its true nature, or whether the membership of particular galaxies to this MS implies the existence of two different modes of star formation. In order to advance our understanding of the MS, here we use a statistically robust Bayesian Spectral Energy Distribution (SED) analysis (CHIBURST) to consistently analyze the star-forming properties of a set of hydro-dynamical simulations of mergers, as well as observations of real mergers and luminous galaxies, both local and at intermediate redshift. We find a very tight correlation between the specific star formation rate (sSFR) of our fitted galaxies, and the typical conditions of the star-forming interstellar medium (ISM), parametrized via a novel quantity: the compactness parameter C, that controls the evolution of dust temperature with time. The normalization of this correlation is bimodal, and such bi-modality relates to the membership of individual galaxies to the MS. As mergers move into the coalescence phase, they increase their compactness and sSFR, creating a scatter in the MS that we measure to be 0.38 dex. The increase in compactness implies that the physical conditions of the ISM smoothly evolve across the MS. One possible interpretation for the slope of the log sSFR- log C correlation is that systems with higher sSFR have smaller physical sizes, whereas the bi-modality between MS objects and outliers suggests the existence of two different regimes of star formation, with distinct ISM conditions.
J. Martinez-Galarza, H. Smith, L. Lanz, et. al.
Wed, 10 Dec 14
42/61
Comments: 21 pages, 11 figures, submitted to ApJ. Comments are welcome