Star formation regulation (or not) by stellar feedback and its varied implications from galactic nuclei to high-redshift galaxies
Abstract: I will present a simple, analytic, quasi-equilibrium theory of star formation regulation by stellar feedback in galactic disks, and argue that this equilibrium explains the observed Kennicutt-Schmidt relation and thus the low efficiency of star formation inferred on galactic scales. I will then discuss conditions in which the quasi-steady balance between star formation and stellar feedback is expected to break down. In these conditions, characterized by short local dynamical times and/or stochastic effects associated with a small number of bright star-forming regions, we expect bursty star formation-gas outflow cycles. I will argue that star formation should be bursty in several observationally and physically interesting environments, including galactic nuclei and high-redshift galaxies, and discuss evidence from both observations and numerical simulations for this phenomenon. I will show that bursty star formation has a variety of important implications, including for the gustiness of galactic winds, the morphological evolution of star-forming galaxies from high redshift to the present day, and the growth of supermassive black holes.