The Environmental Dependence of Galaxy Evolution
Abstract
Observations of galaxy environments have revealed numerous correlations associated
with their intrinsic properties. It is therefore clear that if we are to understand the
processes by which galaxies form and evolve, we have to consider the role of their immediate
environment and how these trends change across cosmic time. In this thesis,
I investigate the relationship between the environmental densities of galaxies and their
associated properties by developing and implementing a novel approach to measuring
galaxy environments on individual galaxy scales with Voronoi tessellations. Using optical
spectroscopy and photometry from GAMA and SDSS, with 250μm far-infrared
observations from the Herschel-ATLAS SDP and Phase-One fields, the environmental
and star formation properties of far-infrared detected and non–far-infrared detected
galaxies are compared out to z ∼ 0.5. Applying statistical analyses to colour, magnitude
and redshift-matched samples, I show there to be significant differences between
the normalised density distributions of the optical and far-infrared selected samples,
at the 3.5σ level for the SDP increasing to > 5σ when combined with the Phase-One
data. This is such that infrared emission (a tracer of star formation activity) favours
underdense regions, in agreement with previous studies that have proposed such a
correlation. I then apply my method to synthetic light cones generated from semianalytic
models (SAMs), finding that over the whole redshift distribution the same
correlations between star-formation rate and environmental density are found. However,
as the SAMs restrict the role of ram-pressure stripping, the fact that we find the
same qualitative results may preclude ram-pressure as a key mechanism in truncating
star formation. I also find significant correlations between isothermal dust temperature and environment, such that the coldest sources reside in the densest regions at the 3.9σ
level, indicating that the observed far-infrared emission in these densest regions is the
product of ISM heating by the older stellar populations.
I then extend my analysis to a deeper sample of galaxies out to z ∼ 2.2, combining
near-infrared and optical photometry from the VIDEO and CFHTLS-D1 observations,
cross-matched in colour, magnitude and redshift against 1.4 GHz VLA radio observations.
Across the entire radio sample, galaxies with radio detected emission are
found to reside in more overdense environments at a 4.0σ significance level. I then divide
my radio sample to investigate environmental dependence on both radio detected
star-forming galaxies and radio detected AGN individually, based upon a luminosity
selection defined as L = 1023 W Hz−1. The same trends with environment are shown
by my Radio-AGN sample (L > 1023 W Hz−1) which favour overdense regions at the
4.5σ level, suggestive of the interaction processes (i.e. major mergers) that are believed
to trigger accretion, in agreement with earlier work that has suggested such a
relationship. At lower radio luminosities, my Radio-SF sample (L < 1023 W Hz−1)
also display a significant trend towards overdense regions in comparison to my nonradio
detected sample, at the less significant level of 2.7σ. This is suggestive of the
low overall bolometric luminosity of radio emission in star forming galaxies, leading
to only the brightest radio emitting star forming galaxies being observed and a bias
towards overdense regions. This is in addition to the fact that the luminosity selection
used to separate AGN from star forming galaxies is not a perfect selection and open to
AGN contamination in the low-luminosity sample. I conclude that the next generation
of deep radio surveys, which are expected to reach many orders of magnitude deeper
than current observations, will remove radio-loud AGN contamination and allow for
the detection of low-luminosity star forming galaxies via radio emission out to high
redshifts.
This work has allowed for the environments of galaxies to be probed on smallerscales
and across both wider and deeper samples than previous studies. With significant
environmental correlations being returned, this indicates that the established processes
responsible for such trends must have influence on the most local of scales.
Publication date
2013-12-10Published version
https://doi.org/10.18745/th.12572https://doi.org/10.18745/th.12572