Some observations have shown that star formation (SF) correlates tightly with the presence of molecular hydrogen (H2); therefore, it is important to investigate its implication on galaxy formation in a cosmological context. In the present work, we implement a sub-grid model (hereafter H2-SF model) that tracks the H2 mass fraction within our cosmological smoothed particle hydrodynamics code GADGET-3 by using an equilibrium analytic model of Krumholz et al. This model allows us to regulate the SF in our simulation by the local abundance of H2 rather than the total cold gas density, which naturally introduces the dependence of SF on metallicity. We investigate the implications of the H2-SF model on galaxy population properties, such as the stellar-to-halo mass ratio (SHMR), baryon fraction, cosmic star formation rate density (SFRD), galaxy specific SFR, galaxy stellar mass functions (GSMF), and Kennicutt-Schmidt (KS) relationship. The advantage of our work over the previous ones is having a large sample of simulated galaxies in a cosmological volume from high redshift to z = 0. We find that low-mass halos with M DM < 1010.5 M are less efficient in producing stars in the H2-SF model at z ≥ 6, which brings the simulations into better agreement with the observational estimates of the SHMR and GSMF at the low-mass end. This is particularly evident by a reduction in the number of low-mass galaxies at M ≤ 108 M in the GSMF. The overall SFRD is also reduced at high z in the H2 run, which results in slightly higher SFRD at low redshift due to more abundant gas available for SF at later times. This new H2 model is able to reproduce the empirical KS relationship at z = 0 naturally, without the need for setting its normalization by hand, and overall it seems to have more advantages than the previous pressure-based SF model.

Document Type


Publication Date


Notes/Citation Information

Published in The Astrophysical Journal, v. 780, no. 2, article 145, p. 1-20.

© 2014. The American Astronomical Society. All rights reserved.

The copyright holder has granted permission for posting the article here.

Digital Object Identifier (DOI)


Funding Information

This work is supported in part by the National Science Foundation (NSF) grant AST-0807491, National Aeronautics and Space Administration (NASA) under Grant/Cooperative Agreement No. NNX08AE57A issued by the Nevada NASA EPSCoR program, and the President's Infrastructure Award from UNLV. Support for Program number HST-AR-12143.01-A was provided by NASA through a grant from the Space Telescope Science Institute, which is operated by the Association of Universities for Research in Astronomy, Inc. under NASA contract NAS5-26555. This research is also supported by the NSF through the Tera Grid resources provided by the Texas Advanced Computing Center (TACC) and the National Institute for Computational Sciences (NICS). K.N. is grateful to the hospitality and the partial support from the Kavli Institute for Physics and Mathematics of the Universe (IPMU), University of Tokyo, the Aspen Center for Physics, and NSF grant 1066293.