Towards a Unified Model for Star Formation: Forging Order from Randomness

Abstract: Astronomy has entered a new era of survey science in which large statistical samples can be scrutinized with unprecedented resolution. However, moving beyond the study of individual objects has revealed fresh puzzles. Three recent observations of protostars and molecular clouds pose a challenge to current star formation models. First, protostellar outflows appear to be randomly oriented with respect to the local magnetic field. Second, forming binary pairs have uncorrelated orientations. Finally, shells identified in molecular clouds suggest stellar winds are much stronger than models predict. I will present magnetohydrodynamic simulations addressing each of these observations and show how synthetic observations and machine learning are essential tools for dissecting and addressing these puzzles. I will discuss how these three results shed light on the role of magnetic fields in star formation, the origin of stellar multiplicity and the evolution of molecular clouds and, altogether, help pave the way for a unified theory of star formation.