Description:

The need for a comprehensive and predictive model for galaxy formation and evolution has never been greater — new and upcoming facilities will generate enormous amounts of cosmological-scale data, but using it to understand the fundamental physics of dark energy and matter will require advances in how we model such systems. I will outline an effort to do this that brings together researchers working in galaxy formation, cosmology, statistical inference and machine learning. After briefly describing some results of this collaboration, this talk will focus on recent advances related to the interaction of cold clouds in hot outflows. These results have led us to a re-evaluation of how galaxy regulation actually occurs, concluding that supernovae in galaxies do not eject large quantities of mass out of galaxies, but instead generate light but high-specific energy winds that heat the surrounding gas, moving galaxy self-regulation out of the galaxy and into the circumgalactic medium (and beyond). I will describe recent work that builds on these insights from the small-scale physics in order to develop large-scale, cosmological models for galaxy formation and evolution.

Speaker: Greg Bryan, Professor of Astronomy, Columbia University

Host: Professor Giacomo Fragione