Invited Presentation

What do we learn about galaxy formation from simulations with realistic feedback?

Author(s): Xiangcheng Ma

Presenter: Xiangcheng Ma (University of California, Berkeley)

Understanding how the physical properties of the ISM, star formation activities, and the effects of stellar feedback evolve across cosmic time is critical for establishing a coherent picture of galaxy formation and evolution. I will talk about results from the Feedback in Realistic Environments (FIRE) project, which is a library of cosmological zoom-in simulations with explicit treatment of the ISM, star formation, and feedback physics. At z>1, the ISM is mainly supported by turbulence driven by gas inflow, mergers, and stellar feedback. Star formation is highly bursty with strong feedback-driven outflows. At any redshift, rotationally supported disks settle down in halos above a certain mass, which increases with redshift. Present-day Milky Way-mass galaxies settle down below z~1. I will talk about several consequences of the ‘bursty’ phase in galaxies at early times, including but not limited to (1) the diversity of gas-phase metallicity gradients in z>0 galaxies, (2) the efficient formation of bound star clusters (progenitors of present-day globular clusters) at early times, and (3) the difficulty of fueling supermassive black hole seeds in the early universe. By comparing simulations from different groups, I will show that the early-time ‘bursty’ phase in the life of galaxies is not well understood. I will discuss what has converged and what has not yet converged with respect to feedback models and what observations in the upcoming years, especially with JWST, we are expecting for better understanding of galaxy formation physics.