Séminaire par Patrick HENNEBELLE (CEA Saclay)

The multi-scale nature of star formation:

Stars are the building blocks of the Universe. Yet in spite of half a century of continuous and intense efforts, their formation process remains insufficiently understood. As a matter of fact, the two most fundamental quantities regarding the star formation process and beyond, namely the star formation rate and the stellar initial mass function, still do not have consensual explanations. Indeed, the huge difficulty that physicists who modelled star formation are facing, stands from its extreme multi-scale nature, as it typically extends over roughly 10 orders of magnitude of spatial scales, from a galaxy to a stellar radii. Starting from an intermediate scale of 1 kpc and covering up to 3-4 decades of spatial scales, I will present a set of numerical simulations, that incorporate the interstellar physics, including gravity, magnetic field, cooling processes as well as stellar feedback aiming at investigating the star formation process. These simulations fail to reproduce both the Schmidt-Kenicutt relation and the measured stellar initial mass function. I will then propose that the spatial dynamics covered, is insufficient and should include larger and smaller scales to respectively address these two problems. I will then argue that the star formation rate can be explained by considering the large scale turbulent driving induced by the largest galactic scales while the stellar initial mass function, in particular its peak, requires to reach astronomical units and the so-called first hydrostatic core.