2014MNRAS.441..525W

We present an analysis of the role of feedback in shaping the neutral hydrogen (H i) content of simulated disc galaxies. For our analysis, we have used two realizations of two separate Milky Way-like ( ∼ L{sstarf}) discs - one employing a conservative feedback scheme (McMaster Unbiased Galaxy Survey), the other significantly more energetic [Making Galaxies In a Cosmological Context (MaGICC)]. To quantify the impact of these schemes, we generate zeroth moment (surface density) maps of the inferred H i distribution; construct power spectra associated with the underlying structure of the simulated cold interstellar medium, in addition to their radial surface density and velocity dispersion profiles. Our results are compared with a parallel, self-consistent, analysis of empirical data from The H i Nearby Galaxy Survey (THINGS). Single power-law fits (P ∝ kγ) to the power spectra of the stronger feedback (MaGICC) runs (over spatial scales corresponding to ∼ 0.5 to ∼ 20 kpc) result in slopes consistent with those seen in the THINGS sample (γ ∼ -2.5). The weaker feedback (MUGS) runs exhibit shallower power-law slopes (γ ∼ -1.2). The power spectra of the MaGICC simulations are more consistent though with a two-component fit, with a flatter distribution of power on larger scales (i.e. γ ∼ -1.4 for scales in excess of ∼ 2 kpc) and a steeper slope on scales below ∼ 1 kpc (γ ∼ -5), qualitatively consistent with empirical claims, as well as our earlier work on dwarf discs. The radial H i surface density profiles of the MaGICC discs show a clear exponential behaviour, while those of the MUGS suite are essentially flat; both behaviours are encountered in nature, although the THINGS sample is more consistent with our stronger (MaGICC) feedback runs.