ASTROPHYSICS

Introduction

On Earth, the water cycle proceeds by three processes. First, water evaporates from the ocean and lakes. Second, water vapor condenses to form clouds. Third, clouds precipitate droplets of water which fall to the Earth as rain. After the rain falls to the Earth, the water droplets collect and eventually flow to lakes and oceans to be evaporated again. Galaxies cycle their star-forming fuel (called “baryons” in distinction to dark matter) in a similar three-step process (see Figure 1). 

Results and Methods

We have explored the physical mechanisms powering galactic superwinds, considering purely thermal, magnetized, and cosmic-ray accelerated superwinds. SuperMUC-NG has allowed us to perform our studies at unprecedented high resolutions from the galactic disk deep into the circumgalactic medium. We have found a shocking dependence on the phase of the medium with respect to the mechanism driving the outflows. Specifically, we utilized the massively parallelized, Eulerian grid-code FLASH [1] to solve the (two-fluid) (magneto-) hydrodynamic equations, evolving Lagrangian tracer particles to identify the fate of star-forming gas in galactic superwinds. We employ FLASH’s unsplit staggered mesh, a finite-volume, high order Godonuv scheme with constrained transport to ensure divergence-free magnetic fields. We include star formation and feedback using active particles, self-gravity of the gas with a multigrid method to solve the Poisson equation, and the Townsend exact integration scheme for radiative cooling. As such, our simulations performed on SuperMUC-NG constitute the physically most sophisticated set of simulations to probe the fate of cold gas clouds deep into the circumgalactic medium to date. As one would expect, magnetized winds are much more filamentary than purely thermal winds. One would expect cosmic rays, which contribute non-thermal pressure support to gas in a similar manner to magnetic fields. Shockingly, we find that superwinds are composed of a much larger number of tiny cold clouds than even the purely thermal winds! Moreover, we find an increasing number of clouds the faster the cosmic rays are transported along magnetic field lines, even though cosmic rays with fast transport significantly reduce the star formation (and hence supernova) rate in galaxies (see Figure 2). 

We have performed follow-up work zooming in for ultra-high resolution to better understand the fate of nonthermally supported clouds. Studying individual clouds in wind-tunnel simulations (see Figure 3), we have found that nonthermally supported clouds survive much more readily than garden-variety thermal plasma, with the survival markedly increased the larger the degree of nonthermal pressure support [3].

Ongoing Research / Outlook

To better understand the results of our superwind simulations, we are presently performing even higher resolution simulations than the wind-tunnel simulations. Namely, we take an individual cloud from the wind-tunnel simulation and zoom in on the interface region to probe the smallest possible scales relevant to our superwind simulations: individual mixing layers. We have found that the clouds outflowing with cosmic rays have cold ~molecular temperatures whereas the thermal winds accelerate predominantly warm atomic clouds. We have found that molecular clouds mix much more efficiently than atomic clouds ([4], Figure 4), providing further understanding of the physical mechanisms supporting enhanced cloud acceleration for the cosmic ray runs than thermal runs in our superwind simulations.