http://arxiv.org/abs/1310.4178
We study the contribution of galaxies with different properties to the global densities of star formation, atomic and molecular hydrogen as a function of redshift. We use the GALFORM model of galaxy formation and evolution, which is set in the LCDM framework. This model includes a self-consistent calculation of the relation between the star formation rate (SFR), and the molecular (H2) and atomic (HI) hydrogen contents of galaxies. The predicted global SFR density and how much of this is contributed by galaxies with different stellar masses and infrared luminosities are in agreement with observations. Also in agreement with observations is the predicted modest evolution of the HI density at z<3. At z<1, ~70% of the predicted H2 in the universe is locked up in galaxies with SFRs in the range 1-10Msun/yr. Model galaxies with moderately large SFRs, >10Msun/yr, make a contribution that increases with redshift, reaching ~50% at z~2. Current high-redshift galaxy surveys are limited to detect carbon monoxide in galaxies with SFR>~30Msun/yr, which in our model make up, at most, ~20% of the H2 in the universe. This contrasts with the predicted cosmic HI density, which is always dominated by galaxies with low SFRs, <1Msun/yr. In terms of stellar mass, the predicted H2 density is always dominated by massive galaxies, Mstellar>10^10Msun, in contrast with the HI density, which is dominated by low mass galaxies, Mstellar<10^9Msun. In the context of upcoming neutral gas surveys, we suggest that the faint nature of the galaxies dominating the HI content of the universe will hamper the optical and infrared counterpart identification, while for H2, we expect follow up observations of molecular emission lines of already existing optical and infrared galaxy catalogues to be able to uncover the H2 density of the universe.
Date added: Thu, 17 Oct 13