• For the majority of galaxies living in regions within∼1σof the median environ- mental density, downsizing shows little or no dependence on environment. An environmental signal is apparent when the ends of the density distribution are compared. In this case, downsizing in high-density regimes appears moderately accelerated compared to low-density ones, with values of Mtr lower by a factor of ∼2.
• We discuss several possibilities for the origin of downsizing based on our results.
We clearly rule out a scenario in which the incidence of star formation decreases uniformly for galaxies at all masses. The weak density dependence also argues against explanations that rely on the accelerated assembly of structure in dense environments, favoring internal mechanisms instead.
• Through comparisons to the expected behavior of dark matter halos, we argue that the dynamical timescale resulting from the growth of structure is at least 5 times longer in galaxies hosted by halos with logM/M⊙≈ 12.5 (logM∗/M⊙ ≈ 10.8) compared to logM/M⊙ = 14.5 (logM∗/M⊙ ≈ 11.6). Because global dynamical scales are also independent of halo mass at a given redshift, it is suggested that the quenching mechanism is strongly mass-dependent with the potential for different physical processes acting in different mass ranges.
There are two obvious avenues for further studies of downsizing. In a forthcoming paper (Bundy et al., in preparation) we discuss the constraints on merging and the growth of galaxies determined by our observations of the total mass function. This will help dissect the role of merging in driving downsizing. In the near future it will also be possible to chart the incidence of AGN among the galaxy population and compare it to the incidence of star formation to probe the link between quenching and AGN. The significant Chandra follow up observations currently underway in the EGS will make that field particularly exciting for such work. Other efforts from the DEEP2 collaboration will focus on precise measures of the evolving star formation rate (Noeske et al., in preparation) and will detail the dependence of other galaxy properties on specific environments (Cooper et al., in preparation).
Acknowledgments
The Palomar Survey was supported by NSF grant AST-0307859 and NASA STScI grant HST- AR-09920.01-A. Support from National Science Foundation grants 00- 71198 to UCSC and AST 00-71048 to UCB is also gratefully acknowledged. We wish to recognize and acknowledge the highly significant cultural role and reverence that the summit of Mauna Kea has always had within the indigenous Hawaiian commu- nity. It is a privilege to be given the opportunity to conduct observations from this mountain.
Chapter 5
The Merger History of Field Galaxies 1
Using deep infrared observations conducted with the CISCO imager on the Subaru Telescope, we investigate the field-corrected pair fraction and the implied merger rate of galaxies in redshift survey fields with Hubble Space Telescope imaging. In the redshift interval, 0.5 < z < 1.5, the fraction of infrared-selected pairs increases only modestly with redshift to 7% ± 6% at z ∼ 1. This is nearly a factor of three less than the fraction, 22% ± 8%, determined using the same technique on Hubble Space Telescope (HST) optical images and as measured in a previous similar study.
Tests support the hypothesis that optical pair fractions at z ∼ 1 are inflated by bright star-forming regions that are unlikely to be representative of the underlying mass distribution. By determining stellar masses for the companions, we estimate the mass accretion rate associated with merging galaxies. Atz ∼1 we estimate this to be 2×109±0.2 M⊙ galaxy−1 Gyr−1. Although uncertainties remain, our results suggest that the growth of galaxies via the accretion of pre-existing fragments remains as significant a phenomenon in the redshift range studied as that estimated from ongoing star formation in independent surveys.
1Much of this chapter has been previously published as Bundy et al. (2004)
5.1 Introduction
The hierarchical growth of dark matter halos is thought to govern the assembly his- tory and morphological evolution of galaxies. Nearby examples of interacting and merging galaxies are well known, and many attempts to survey the merging and mass accretion rate at various redshifts have been made by several groups (Burkey et al. 1994; Carlberg et al. 1994; Yee & Ellingson 1995; Patton et al. 1997; Le F`evre et al. 2000; Patton et al. 2000, 2002; Conselice et al. 2003). Strong evolution of the global merger rate was used to explain the observed faint galaxy excess (Broadhurst et al. 1992), the evolution of the luminosity function (Lilly et al. 1995a; Ellis et al.
1996), and that of galaxy morphologies (Giavalisco et al. 1996; Brinchmann et al.
1998). Evolution of the merger rate can also be used to place constraints on structure formation (Baugh et al. 1996; Kauffmann 1996).
Le F`evre et al. (2000) used Hubble Space Telescope (HST) F814W images of redshift survey fields to measure the pair fraction to z ∼ 1. They found an increase in the field-corrected pair fraction to 20% at z ∼ 0.75-1. However, as Le F`evre et al. (2000) discuss, various biases affect this result. For example, in the restframe blue, bright star-forming regions, possibly triggered by interactions, might inflate the significance of pair statistics and give a false indication of the mass assembly rate.
Infrared observations are less biased by star formation and serve as a better tracer of the underlying stellar mass in galaxies (Broadhurst et al. 1992). Dickinson et al.
(2003) employed this in their investigation of the global stellar mass density forz <3.
They find that 50 to 70% of the present-day stellar mass was in place byz ∼1. A sec- ond line of evidence, the decline fromz ∼1 to 0 in the global star formation rate (e.g., Lilly et al. 1996), provides further support for the contention that galaxy growth was not yet complete atz ∼1. Though a chronological picture of stellar mass assembly is emerging, the processes driving it are not understood. Are star-formation and stellar mass assembly induced mainly through the gradual accretion of gas converted quies- cently into stars, or does assembly occur through merging, potentially accompanied by tidally-induced star formation? Characterizing the continued growth of galaxies,
and specifically the contribution from galaxy mergers since z ∼1, is the major goal of this work.