Simulating the Formation of Massive Star Clusters

Abstract: Star cluster formation is a challenging multi-physics problem to simulate: stars form in molecular clouds in which gravity, supersonic MHD turbulence, radiative cooling and heating, ISM chemistry, and stellar feedback all have a role to play. I will present a novel numerical approach to this problem adapted from the Feedback in Realistic Environments (FIRE) simulations. I will describe these physics methods and their implications for cluster formation in turn. I will then present the results of a suite of cloud collapse simulations that survey a wide range of ISM conditions to elucidate the connection between galactic environments and the star clusters that form in them. Due to stellar feedback, the star formation efficiency (SFE) of a virialized cloud is mainly an increasing function of its surface density, similar to the trend observed between local molecular clouds and dense clumps. The mass of bound star clusters surviving the embedded phase is subsequently a function of this SFE, and an initial cluster mass function dN/dM \~ M^{-2} emerges from the dynamics of turbulent fragmentation. The bound star clusters formed in the simulations are remarkably similar in shape to observed young massive clusters; I will show that this structure emerges naturally from the process of hierarchical assembly from fragmented subclusters.