Volume 390, Issue 4 pp. 1349-1371

Damped Lyman α systems in galaxy formation simulations

Andrew Pontzen

Corresponding Author

Andrew Pontzen

Institute of Astronomy, Madingley Road, Cambridge CB3 0HA

E-mail: [email protected]Search for more papers by this author
Fabio Governato

Fabio Governato

Astronomy Department, Box No. 351580, University of Washington, Seattle, WA 98195, USA

Search for more papers by this author
Max Pettini

Max Pettini

Institute of Astronomy, Madingley Road, Cambridge CB3 0HA

Search for more papers by this author
C. M. Booth

C. M. Booth

Department of Physics, Institute for Computational Cosmology, University of Durham, South Road, Durham

Sterrewacht Leiden, University of Leiden, P.O. Box 9513, 2300 RA Leiden, the Netherlands

Search for more papers by this author
Greg Stinson

Greg Stinson

Astronomy Department, Box No. 351580, University of Washington, Seattle, WA 98195, USA

Department of Physics and Astronomy, McMaster University, Hamilton, ON L8S 4M1, Canada

Search for more papers by this author
James Wadsley

James Wadsley

Department of Physics and Astronomy, McMaster University, Hamilton, ON L8S 4M1, Canada

Search for more papers by this author
Alyson Brooks

Alyson Brooks

Astronomy Department, Box No. 351580, University of Washington, Seattle, WA 98195, USA

Search for more papers by this author
Thomas Quinn

Thomas Quinn

Astronomy Department, Box No. 351580, University of Washington, Seattle, WA 98195, USA

Search for more papers by this author
Martin Haehnelt

Martin Haehnelt

Institute of Astronomy, Madingley Road, Cambridge CB3 0HA

Search for more papers by this author
First published: 27 October 2008
Citations: 38

ABSTRACT

We investigate the population of z= 3 damped Lyman α systems (DLAs) in a recent series of high-resolution galaxy formation simulations. The simulations are of interest because they form at z= 0 some of the most realistic disc galaxies to date. No free parameters are available in our study: the simulation parameters have been fixed by physical and z= 0 observational constraints, and thus our work provides a genuine consistency test. The precise role of DLAs in galaxy formation remains in debate, but they provide a number of strong constraints on the nature of our simulated bound systems at z= 3 because of their coupled information on neutral H i densities, kinematics, metallicity and estimates of star formation activity.

Our results, without any parameter tuning, closely match the observed incidence rate and column density distributions of DLAs. Our simulations are the first to reproduce the distribution of metallicities (with a median of ZDLA≃ Z/20) without invoking observationally unsupported mechanisms such as significant dust biasing. This is especially encouraging given that these simulations have previously been shown to have a realistic 0 < z < 2 stellar mass–metallicity relation. Additionally, we see a strong positive correlation between sightline metallicity and low-ion velocity width, the normalization and slope of which come close to matching recent observational results. However, we somewhat underestimate the number of observed high-velocity width systems; the severity of this disagreement is comparable to other recent DLA-focused studies.

DLAs in our simulations are predominantly associated with dark-matter haloes with virial masses in the range 109 < Mvir/M < 1011. We are able to probe DLAs at high resolution, irrespective of their masses, by using a range of simulations of differing volumes. The fully constrained feedback prescription in use causes the majority of DLA haloes to form stars at a very low rate, accounting for the low metallicities. It is also responsible for the mass–metallicity relation which appears essential for reproducing the velocity–metallicity correlation. By z= 0, the majority of the z= 3 neutral gas forming the DLAs has been converted into stars, in agreement with rough physical expectations.

The full text of this article hosted at iucr.org is unavailable due to technical difficulties.