Interpreting the Properties of Galaxies

Galaxies exhibit a wide range of physical properties (e.g., luminosities, colors, velocity widths, star formation, gas and stellar content) and the evolutionary processes responsible for these properties are numerous and complex. Understanding which processes shape the observable properties of galaxies and which others play only a minor role, inherently requires a large sample of galaxies. Moreover, if we want to understand why galaxies have the properties they do, we need a theory of galaxy formation. The standard paradigm of galaxy formation assumes that most of the matter is dark and dissipationless and that, under the influence of gravity, structures on galactic and larger scales grow hierarchically (from Gaussian initial conditions) with smaller objects forming first. Gas, moving under the gravitational influence of the dark component, dissipates and collapses at the center of the potential wells provided by the dark matter. In this picture the internal structure of the dark matter clumps and their formation history regulate the global properties of galaxies. However, these properties must also depend on how gas cools to form the dense clouds that seed star formation and how star formation affects the surrounding medium with the injection of energy and heavy elements. I show how simple, ``semi-analytic'' parameterizations are used to describe the highly non-linear aforementioned processes and to predict a wide range of properties of the galaxy population for any specific cosmogony. I then present a simple and flexible framework to extract from the numerous observable properties of disk galaxies that semi-analytic models predict, only those that are needed to characterize the sample as a whole. This framework makes use of the well-know statistical technique of Principal Component Analysis (PCA). Moreover, I correlate the semi-analytic assumptions with the PCA findings and determine which, among our theoretical assumptions, shape the observable galaxies' properties. When applied to large dataset of observable properties of galaxies, such as the upcoming Sloan Digital Sky Survey, this framework will provide useful insights on the process of galaxy formation. I also measure the evolution of galaxy sizes in the Hubble Deep Field North (HDFN), a rich dataset which contains a large number of galaxies at high redshifts. I adopt as the angular size estimator the radius within which half of the total galaxy light is contained. Galaxy magnitudes are computed based upon the Petrosian metric radius which is relatively insensitive to redshift, making it a good probe of evolutionary changes in the galaxy size. I find that the angular size distribution of galaxies in the HDFN is strongly peaked at very small sizes (~0.2 arcsec). In order to study the evolution of galaxy sizes I use published photometric redshifts and construct volume-limited samples out to depths of z=1,2, and 3. I find that the mean physical radius of galaxies in the HDFN exhibits no significant evolution in the redshift range from z=3 to z=0.4. Finally, I make use the HDFN to look for high redshift quasars. Quasars are believed to be the visible manifestation of the accretion of matter onto supermassive black holes and, being among the most distant and luminous objects in the universe, hold a primary role as cosmological probes. As such they hold important clues on the process of galaxy formation. Due to their stellar-like nature, quasars cannot be distinguished from stars in single images of the sky, as galaxies can. However, multi-color selection techniques have proven very successful in selecting quasar candidates at high redshifts. Moreover, to make the best use of the depth of the HDFN, I developed a morphological technique for identifying quasars in the HDFN which complements the color technique. I find one quasar candidate, 7 pointlike objects with colors consistent with quasars or stars, 18 stars, and 15 slightly resolved objects, 12 of which have colors consistent with quasars or stars.