This thesis explores the evolution of massive galaxies (M * > 1011M ʘ) by conducting the largest multiple-component Sersic light-profile fitting study to date of the rest-frame optical and ultra-violet morphologies of galaxies at redshifts 1 < z < 3. Despite many of the recent advances in galaxy formation and evolution models, the physical processes which are responsible for driving morphological transformations and star-formation quenching remain unclear. By undertaking a detailed study of the individual bulge and disk components of these massive systems, the work presented in this thesis addresses these outstanding issues by exploring not only how the sizes of the individual components evolve with redshift, but also how the overall bulge and disk fractions evolve, and how these trends are connected to star-formation quenching of the separate components. In order to perform this analysis, I have combined the latest high-resolution near-infrared HST WFC3/IR and ACS imaging provided by the CANDELS survey in the UDS and COSMOS fields and have presented a robust procedure for morphological multiple-component Sersic light-pro le model fitting across the 0:6μ m to 1:6μ m wavelength range sampled by CANDELS. This procedure is discussed in depth along with the tests I have undertaken to assess its reliability and accuracy. This approach has enabled me to generate separate bulge and disk component model photometry, allowing me to conduct individual component SED fitting in order to determine decomposed stellar-mass and star-formation rate estimates for the separate bulge and disk components. The results presented in this work reveal that the sizes of the bulge and disk components lie both on and below the local size-mass relations, confirming that the size evolution required by the previously reported compact sizes of high-redshift galaxies extends to both galaxy components. However, I find evidence that the bulge components display a stronger size evolution with redshift than the disks as, at 1 < z < 3, the bulges are a median factor of 3:09 ± 0:2 times smaller than similarly massive local early-type galaxies, whereas the disks are a median factor of 1:77 ± 0:1 times smaller than similarly massive local late-type galaxies. By including decomposed star-formation rates for the individual bulge and disk components, this work also reveals that while the growth of individual components through, for example, inside-out processes such as minor merging, are consistent with the size evolution of these systems, the addition of larger newly quenched systems to the galaxy population, for the disk components at least, may also play an important role in the observed size evolution of massive galaxies. By exploring the evolution of the bulge and disk-dominated fractions with redshift, I find that 1 < z < 3 marks a key transition era in cosmic time where these most massive galaxies appear to be undergoing dramatic structural transformations. Within this redshift range there is a decline in the population of disk-dominated galaxies and a gradual emergence of increasingly bulge-dominated systems. However, despite the rise of S0-type galaxies, even by z = 1 I do not yet find a significant fraction of "pure" bulges comparable to the giant ellipticals which comprise the majority of the local massive galaxy population. In addition to studying how the overall bulge and disk dominated fractions evolve with redshift, by incorporating the star-formation rate and stellar-mass estimates for the separate components and imposing new, highly conservative criteria, I con firm that a significant fraction of passive galaxies are disk-dominated (18± 5%) and a significant fraction of star-forming galaxies are bulge-dominated (11 ±4%). The presence of passive disks and star-forming bulges has interesting implications for the models of galaxy evolution as they suggest that the processes which quench star-formation may be distinct from the mechanisms which cause morphological transformations. Finally, the detailed morphological analysis presented in this work has also allowed me to explore the axial ratio distributions of these most massive high-redshift galaxies, which provides additional insight into the structure of the passive and star-forming bulge and disk-dominated sub-populations. Whilst the overall axial ratio distributions for star-forming disks are peaked, I find tentative evidence that the largest and most active star-forming disks are flatter. I have also been able to further demonstrate that by selecting the most active star-forming disks and comparing to extreme star-forming (sub-)mm selected galaxies, the axial ratio distributions of the two samples appear to be comparably flat, thus reconciling the observed structures of these populations.

No description available.

No description available.

No description available.

No description available.

No description available.

I use observational evidence to examine the nature and role in galaxy evolution of a population of dust-obscured galaxies (DOGs) at z ∼ 2. These objects are selected with the Spitzer Space Telescope, are bright in the mid-infrared (mid-IR) but faint in the optical, and contribute a significant fraction of the luminosity density in the universe at z ∼ 2. The first component of my thesis is a morphological study using high spatial resolution imaging with the Hubble Space Telescope of two samples of DOGs. One set of 33 DOGs have mid-IR spectral features typical of an obscured active galactic nucleus (AGN) (called power-law DOGs), while the other set of 20 DOGs have a local maximum in their spectral energy distribution (SED) at rest-frame 1.6μm associated with stellar emission (called bump DOGs). The host galaxy dominates the light profile in all but two of these DOGs. In addition, bump DOGs are larger than power-law DOGs and exhibit more diffuse and irregular morphologies; these trends are consistent with expectations from simulations of major mergers in which bump DOGs evolve into power-law DOGs. The second component of my thesis is a study of the dust properties of DOGs, using sub-mm imaging of 12 power-law DOGs. These power-law DOGs are hyper- luminous (2 × 10¹³ L⊙) and have predominantly warm dust (T(dust) > 35 - 60 K). These results are consistent with an evolutionary sequence in which power-law DOGs represent a brief but important phase when AGN feedback heats the interstellar medium and quenches star-formation. The third component of my thesis is a study of the stellar masses and star- formation histories of DOGs, using stellar population synthesis models and broad- band photometry in the rest-frame ultra-violet, optical, and near-IR. The best-fit quantities indicate bump DOGs are less massive than power-law DOGs. The relatively low stellar masses found from this line of analysis favor a merger-driven origin for ULIRGs at z ∼ 2.

No description available.

No description available.

No description available.