Luminosity functions and galaxy bias in the Sloan Digital Sky Survey

  • James G. Cresswell

Student thesis: Doctoral Thesis


With the wealth of data provided by modern imaging and redshift surveys, it has become essential to obtain measurements of galaxy properties in an automated fashion, and analyse them in a statistical manner. In this thesis we focus on two fundamental characteristics of the galaxy populations, their distributions in luminosity and their spatial clustering signal. We examine galaxies from the SDSS-II galaxy redshift survey, quantifying their distribution in luminosity as a function of rest-frame colour and visual morphology taken from the Galaxy Zoo project. In forming samples of the galaxies selected by colour, morphology and luminosity we observe that there are significant populations of both red spiral galaxies and blue early-type galaxies at redshifts of 0.01 < z < 0.2. To aid us in the description and analysis of these distributions we fit them with a parametric model, the well known Schechter function, in which we include parameterisations for the redshift evolution of absolute magnitude and number density. In comparing the results of fitting our various samples with the function we find that blue galaxies exhibit significantly different behaviour to spiral galaxies and that red galaxies exhibit significantly different behaviour to early-type galaxies. We then go on to measure the 2-point clustering statistics of the galaxies in order to explore the dependence of galaxy clustering bias on luminosity and colour. We fit our measurements of scale dependent galaxy clustering with several phenomenological models of galaxy bias. This allows us to describe how the large-scale bias and the transition to non-linear clustering behaviour depend on luminosity and colour. Our key result from this investigation is that red and blue galaxies exhibit significantly different clustering behaviour in both the amplitude of their large scale bias and their non-linear bias contributions at smaller scales. In conclusion we find that in order to extract the maximum amount of cosmological information from future surveys they must be planned and analysed with the understanding that galaxy clustering depends strongly on both luminosity and colour, and that galaxy morphology contains considerable information about a galaxies formation history, independent of its colour.
Date of AwardMay 2010
Original languageEnglish
Awarding Institution
  • University of Portsmouth
SupervisorWill Percival (Supervisor) & Bob Nichol (Supervisor)

Cite this