The IllustrisTNG simulations are some of the most advanced hydrodynamical, large-volume cosmological simulations to date. While these simulations predict countless properties of the stars, black holes, and gas in galaxies, they do not currently compute the abundance of the different phases of neutral hydrogen: atomic and molecular (HI and H2). In Diemer et al. 2018, we post-processed the Illustris and IllustrisTNG simulations to compute their abundances in galaxies (see also the research page).
The resulting data are publicly available as part of the IllustrisTNG data release. Once you have created a user account, you can click on a simulation (e.g., “TNG100-1”). At the bottom of the simulation page, the available supplementary datasets are listed. If “Molecular and atomic hydrogen (HI+H2) galaxy contents” is listed there, you can obtain the HI/H2 data using the download link. Alternatively, you can find the same files here.
In Diemer et al. 2019, we compare Illustris and IllustrisTNG 100/300 to HI and H2 observations. To avoid overcrowding the figures in the paper, we omitted the original Illustris-1 simulation from the plots. Below, you can download complete sets of figures that include TNG50, that compare Illustris TNG100 with Illustris-1, and that compare TNG100 with its lower-resolution cousin, TNG100-2, to highlight any issues with convergence.
From the first figure set, it is apparent that TNG50 is very compatible with TNG100/300, adding a nice convergence test at higher resolution. Moreover, TNG50 matches some observations well (e.g., the HI mass function). From the second figure set, we conclude that the HI and H2 abundances in Illustris-1 match the observations much more poorly than IllustrisTNG. Thus, the latter is the preferred simulation suite for any work related to HI and H2.
Compared to the original paper, these figure sets are based on updated data files that fix a few bugs that have a minor impact on the results. Moreover, the TNG300 dataset was extended to lower stellar masses. If you have any questions about the data or these figures, please do not hesitate to contact me.