Supermassive black holes (BHs) and their host-galaxies are thought to evolve in tandem, with the energy output from the rapidly-accreting BH regulating star formation and the growth of the BH itself. The goal of better understanding this process has led to much work focussing on the properties of quasars at high redshifts, $z\gtrsim 2$, when cosmic star formation and BH accretion both peaked. At these redshifts, however, ground-based statistical studies of the quasar population generally have no access to the rest-frame optical spectral region, which is needed to measure H$\beta$-based BH masses and narrow line region outflow properties. The cornerstone of this thesis has been a new near-infrared spectroscopic catalogue providing rest-frame optical data on 434 luminous quasars at redshifts $1.5 \lesssim z \lesssim 4$. At high redshift, $z \gtrsim 2$, quasar BH masses are derived using the velocity-width of the CIV broad emission-line, based on the assumption that the observed velocity-widths arise from virial-induced motions. However, CIV exhibits significant asymmetric structure which suggests that the associated gas is not tracing virial motions. By combining near-infrared spectroscopic data (covering the hydrogen Balmer lines) with optical spectroscopy from SDSS (covering CIV), we have quantified the bias in CIV BH masses as a function of the CIV blueshift. CIV BH masses are shown to be over-estimated by almost an order of magnitude at the most extreme blueshifts. Using the monotonically increasing relationship between the CIV blueshift and the mass ratio BH(CIV)/BH(H$\alpha$) we derive an empirical correction to all CIV BH-masses. The correction depends only on the CIV line properties and therefore enables the derivation of un-biased virial BH mass estimates for the majority of high-luminosity, high-redshift, spectroscopically confirmed quasars. Quasars driving powerful outflows over galactic scales is a central tenet of galaxy evolution models involving 'quasar feedback' and significant resources have been devoted to searching for observational evidence of this phenomenon. We have used [OIII] emission to probe ionised gas extended over kilo-parsec scales in luminous $z\gtrsim2$ quasars. Broad [OIII] velocity-widths and asymmetric structure indicate that strong outflows are prevalent in this population. We estimate the kinetic power of the outflows to be up to a few percent of the quasar bolometric luminosity, which is similar to the efficiencies required in recent quasar-feedback models. [OIII] emission is very weak in quasars with large CIV blueshifts, suggesting that quasar-driven winds are capable of sweeping away gas extended over kilo-parsec scales in the host galaxies. Using data from a number of recent wide-field photometric surveys, we have built a parametric SED model that is able to reproduce the median optical to infrared colours of tens of thousands of AGN at redshifts $1 < z < 3$. In individual objects, we find significant variation in the near-infrared SED, which is dominated by emission from hot dust. We find that the hot dust abundance is strongly correlated with the strength of outflows in the quasar broad line region, suggesting that the hot dust may be in a wind emerging from the outer edges of the accretion disc.

Philosophiae Doctor - PhD / There have been a number of attempts to measure the expansion rate of the Universe using age-dating of Luminous Red Galaxies (LRGs). Assuming that stars in LRGs form at the same time, age-dating of two populations of LRGs at different redshifts can provide an estimate of the time different associated with the corresponding redshift interval (dz/dt). This gives a direct estimate of the Hubble parameter H (z) at the average redshift of the two populations. In this thesis, we explore the validity of this method by using two different sets of data. Firstly, we select a homogeneous sample of passively evolving galaxies over 0.10 < z < 0.40 from the Sloan Digital Sky Survey Data Release Seven (SDSS-DR7) catalogue by applying a refined criteria, which is based on absolute magnitude. Secondly, we carry out series of observations on the Southern African Large Telescope (SALT) to obtain spectra of LRGs at two narrow redshift ranges z ' 0.40 and z ' 0.55 in order to calculate the Hubble parameter H(z) at z ' 0.47. We utilise two distinct methods of age-dating including the use of absorption Lick index lines and full spectral fitting on high signal-to-noise galaxy spectra from our sample. By establishing the age-redshift relation of the quiescent, passively evolving galaxies from SDSS, we obtain three improved new observational H(z) data points which are H(z) = 76.8 5.3 km s􀀀1Mpc􀀀1 at z ' 0.28, H(z) = 78.5 6.8 km s􀀀1Mpc􀀀1 at z ' 0.30 and H(z) = 86.3 7.6 km s􀀀1Mpc􀀀1 at z ' 0.32 respectively. We also find another H(z) value of 105 39 km s􀀀1Mpc􀀀1 at z ' 0.47 when age-dating LRGs observed with SALT. Combining all 4 data points with another 25 data points in the literature, we place better constraints on cosmological models and find the matter density parameter to be constrained by m = 0:32+0:05 􀀀0:06 and the Hubble constant to be H0 =68.5 2.4. These results are very consistent with other studies. Through this work, we are able to demonstrate that the cosmic chronometers approach can potentially be used to explore the evolution of the Universe.

This thesis presents a detailed study of the impact of radio-AGN on star formation and the interstellar medium (ISM) of galaxies across cosmic time. To do this, this thesis uses far-IR/sub-mm data from the Herschel Space Observatory. I create a well-selected sample of 1599 radio sources using the NRAO VLA Sky Survey (NVSS) and Faint Images of the Radio Sky at Twenty-Centimeters (FIRST) data in combination with the UKIRT Infrared Deep Sky Survey - Large Area Survey (UKIDSS - LAS) data. I find that the far-IR luminosities and dust temperatures of radio galaxies are lower in comparison to those of non-radio-detected galaxies. This luminosity deficit grows with increasing stellar mass. I argue that the reasons for these differences is probably due to indirect radio-AGN feedback, i.e. radio jets mechanically heat the halo-environment, preventing external sources of cold gas from entering the host and forming stars. The far-IR luminosity and dust temperature is found to decrease as a function of radio source size. I find the most likely explanation for this is jet-induced star formation while the jets are confined to the ISM. Finally, a method for identifying reliable high-z, star-bursting radio sources in the H-ATLAS is described with which statistically significant studies of radio-jet induced star formation may be undertaken.

The stellar populations harboured by some of the Universe’s earliest galaxies are within observational reach. Determining the details of these stellar populations and their formation histories within the first billion years after the Big Bang is crucial for both understanding the earliest stages of galaxy evolution and for assessing the contribution of early star-forming galaxies to cosmic reionization. This thesis presents observational measurements of the rest-frame UV and optical colours of star-forming Lyman Break galaxies (LBGs) at redshifts 4 < z < 9, and their inferred stellar population parameters. By combining ground-based ~1 deg² surveys with deeper, narrower space-based deep-field surveys, we have constrained the rest-frame UV spectral slope of galaxies over a wide-range of cosmic time (4 < z < 9) and luminosity (−23 < MUV < −17) in a self-consistent way. To do so, we developed simulations to allow the inference of intrinsic colours from noisy, potentially biased observations. With these simulations, a robust UV colour measurement method was devised in preparation for the Hubble Ultra Deep Field 2012 (UDF12) survey. Then, after delivery of the UDF12 data, our technique and simulations were applied to yield the first bias-free measurements of the UV spectral slope of galaxies at z ≈ 7 and 8. We found no support for the previously claimed dominant sub-population of exotically blue, faint galaxies at z ≈ 7. In fact with careful consideration of their errors and selection biases, even the most extreme galaxies we observed can have their colours explained by stellar population synthesis models of unremarkable parameters. Expanding this study to brighter, rarer, galaxies required the inclusion of wide-area ground-based survey data, and consequently a more focused examination of galaxies at z ≈ 5. We selected high signal-to-noise galaxies from four fields, with absolute magnitudes spanning MUV = −22.5 to −17.5, and measured their rest-frame UV spectral slopes. Coupling these measurements with our simulated observations, we were able to determine the width of the intrinsic colour distribution of galaxies at z ≈ 5. We found that brighter galaxies are not only on average redder than their fainter counterparts, but they are also less self-similar in their colours. The redder average UV colours of brighter galaxies can be attributed to those galaxies being either older, or more dust reddened. By pairing these measurements, which are primarily a probe only of the presently forming portion of the stellar population, with those of LBG’s Balmer Breaks, which are more sensitive to bygone star formation, we were able to break this age–dust degeneracy and conclude that, at z ≈ 5, brighter galaxies are more heavily reddened than fainter galaxies even though their stars are no older.

We present new optical and radio spectroscopic observations of the remarkable galaxy Malin 1. This galaxy has unique features that include an extremely low surface brightness disk with an enormous mass of neutral hydrogen, and a low luminosity Seyfert nucleus. Malin 1 is exceptional in its values of MHO, LB, and MHI /Ln, and modest in its surface mass density of gas and stars. Spirals with large Min /LB tend to have low mean column densities of HI, and are close to the threshold for star formation due to instabilities in a rotating gas disk. In these terms, Malin 1 has a disk with extremely inefficient star formation. The bulge spectrum is dominated by the absorption features of an old, metal rich stellar population, although there is some evidence for hot (young) stars. The emission line excitations and widths in the nucleus are typical of a Seyfert galaxy; but Malin 1 is in the lowest 5% of the luminosity function of Seyferts, despite a copious fuel supply. Malin 1 is in a low density region of the universe. We propose it as an unevolving disk galaxy, where the surface mass density is so low that the chemical composition and mass fraction in gas change very slowly over a Hubble time. Its properties are similar to those of the damped Lyman -a absorption systems seen in the spectra of high redshift quasars. We emphasize that there are strong observational selection effects against finding gas -rich galaxies that are both massive and diffuse. Finally, we suggest that large and massive HI disks may have formed as early as z - 2, and remained quiescent to the present day. Subject headings : individual (Malin 1) - galaxies : photometry - galaxies : Seyfert - galaxies : stellar content - radio sources : 21 cm radiation - stars : formation

A study of the Galactic Center stellar population is continuing with a sensitive 2μm CCD camera. Using a 64 x 64 detector array, background limited images are recorded with modest amounts of observing time (tob, 20 sec to reach K =13). Magnitudes have been extracted using DAOPHOT from repeated imaging of the central 5' x 5' to search among approximately 1500 stars for long period variables (LPV's, P > 200d), particularily Miras. Miras have a well defined period - luminosity relationship as well as one in period -mass. This program investigates the nature of highly luminous stars at the Galactic Center. Presently 12 variables have been found and have several characteristics consistant with Miras. They have a maximum bolometric luminosity of -4.4 mag which supports the case that high luminosity stars in the central 6 pc are young supergiants.

No description available.

In this thesis, I present the results of a new search for, and study of, luminous galaxies in the first billion years of cosmic time. This work is primarily based on a new selection of bright (L≫L*) Lyman-break galaxies (LBGs) at z ≅6 and z≅7 in the UltraVISTA first and second data releases (DR1, DR2) and the UKIDSS (UKIRT Infrared Deep Survey) UDS DR10 (Ultra Deep Survey). The UltraVISTA survey provides deep Y, J,H andKs near-imaging over 1.5 deg² of the Cosmic Evolution Survey (COSMOS) field and the UKIDSS UDS provides J,H and K band data overlapping with the Subaru XMM-Newton Deep Survey (SXDS), with both fields also containing deep optical and mid-infrared imaging essential for the clean detection of z > 5 galaxies. The fields combined provide an unprecedented 1.65 deg² of deep multiwavelength data with which to securely select LBGs using a photometric redshift fitting technique, which can additionally remove probable low-redshift galaxy interlopers and galactic dwarf stars that can contaminate ground-based samples. At z ≅7, the DR1 of the UltraVISTA survey was used to select a sample of ten high-redshift galaxy candidates, which extended to a 5σ limiting magnitude of Y + J ~ 25 (AB magnitude, 2-arcsec diameter circular aperture) over 1 deg². A stack of the four most robust objects from the sample indicated that they were massive (M*≅5 × 109M ʘ), had blue rest-frame UV slopes (β ≅−2.0±0.2) and were highly star-forming (SFR ≅25–50Mʘ yr−1) when compared to previous, fainter, samples of galaxies at z = 7. The number counts of z≅7 galaxies selected within the UltraVISTA DR1 survey was higher than that expected from extrapolations of the rest-frame UltraViolet (UV) luminosity function (LF) from fainter data, a result that was strongly confirmed with an improved search for z ≅7 galaxies using the UltraVISTA DR2 imaging and the UDS field. A total of 34 galaxies at 6.5 < z < 7.5 were found in the combined fields, which included the previously identified robust galaxies from the DR1 imaging. This expanded sample allowed the first determination of the rest-frame UV LF in the range −23.0 < MUV < −21.5 at z ≅7, and the results reveal a power-law decline to bright magnitudes in contrast to the commonly assumed exponentially declining Schechter function extrapolated from fainter data. The excess of galaxies observed at bright magnitudes cannot be accounted for by gravitational lensing or by significant contamination of the sample by Active Galactic Nuclei (AGN) . The observed LF is well described by a double power law, which at the bright end follows the form of the underlying dark matter halo mass function, suggesting that the physical mechanism that inhibits star formation activity in massive galaxies (e.g. AGN feedback or some other form of ‘mass quenching’) has yet become efficient at z ≅7. The deeper imaging data confirm that the z ≅7 LBGs show blue rest-frame UV slopes (median β = −2.0) and are massive (up to M*≅1010M ʘ). Furthermore, an analysis of the ground-based imaging shows that the majority are resolved consistent with larger sizes (r1/2 ≅1–1.5 kpc) than displayed by less massive galaxies. Finally, a new search for z ≅6 galaxies within the UltraVISTA and UDS datasets was undertaken, resulting in a sample of 266 LBGs (−22.7 < MUV < −20.5) galaxies with which to investigate the rest-frame UV LF. The potential contamination by galactic brown dwarfs was investigated quantitatively using a simple model of the Galaxy, showing that the expected contamination rate of the sample was < 3 per cent, and that the stars can be effectively removed by fitting standard stellar spectra to the observed photometry. The galaxy surface density in the UltraVISTA/COSMOS field exceeds that in the UDS/SXDS by a factor of ≅1.4, indicating strong cosmic variance between the two fields. The number counts of galaxies we find are a factor of 2 lower than predicted by the recent LF determination by Bouwens et al., and the derived rest-frame UV LF at z ≅6 revealed that an under dense UDS field can account for some of the observed differences between previous analyses. An evolution in the characteristic magnitude between z ≅5 and z≅7 of ∆M*~ 0.5 was found in contrast to other smaller area surveys, and a double power law was shown to equally well describe the LF at z = 6 as compared to the commonly assumed Schechter function. The bright-end of the LF at z ≅6 tentatively shows a steeper decline than found at z ≅7, which could indicate the onset of mass quenching of the most massive galaxies or the rise of dust obscuration. Comparison with the predictions of the latest theoretical models and simulations of galaxies reveals that most models require substantial (A1500 ~ 1.5–2) average dust extinction at the bright end to reproduce the shape of the galaxy UV LF at z ≅7.

This thesis documents the hydrodynamics code Cholla and a numerical study of multiphase galactic winds. Cholla is a massively-parallel, GPU-based code designed for astrophysical simulations that is freely available to the astrophysics community. A static-mesh Eulerian code, Cholla is ideally suited to carrying out massive simulations (> 2048 ³ cells) that require very high resolution. The code incorporates state-of-the-art hydrodynamics algorithms including third-order spatial reconstruction, exact and linearized Riemann solvers, and unsplit integration algorithms that account for transverse fluxes on multidimensional grids. Operator-split radiative cooling and a dual-energy formalism for high mach number flows are also included. An extensive test suite demonstrates Cholla's superior ability to model shocks and discontinuities, while the GPU-native design makes the code extremely computationally efficient - speeds of 5-10 million cell updates per GPU-second are typical on current hardware for 3D simulations with all of the aforementioned physics. The latter half of this work comprises a comprehensive study of the mixing between a hot, supernova-driven wind and cooler clouds representative of those observed in multiphase galactic winds. Both adiabatic and radiatively-cooling clouds are investigated. The analytic theory of cloud-crushing is applied to the problem, and adiabatic turbulent clouds are found to be mixed with the hot wind on similar timescales as the classic spherical case (4-5 t_cc) with an appropriate rescaling of the cloud-crushing time. Radiatively cooling clouds survive considerably longer, and the differences in evolution between turbulent and spherical clouds cannot be reconciled with a simple rescaling. The rapid incorporation of low-density material into the hot wind implies efficient mass-loading of hot phases of galactic winds. At the same time, the extreme compression of high-density cloud material leads to long-lived but slow-moving clumps that are unlikely to escape the galaxy.

La population de galaxies poussiéreuses ayant un fort taux de formation stellaire (Dusty Star Forming Galaxies, DSFGs) joue un rôle très important dans l’histoire de l’univers, avec des taux de formation d’étoiles allant de quelques centaines à quelques milliers de masses solaires par an. Les sondages infrarouges, comme ceux entrepris à l’aide du satellite Herschel, nous offrent l’opportunité de recenser de manière approfondie ces DSFGs jusqu’à de grands décalages spectraux. Cependant, jusqu’à présent seul un petit nombre de DSFG détecté par Herschel ont été confirmés pour être à des décalages spectraux supérieurs à 4. Les modèles de formation et d’évolution des galaxies stipulent généralement que la population de DSFG à z > 4 sont les progéniteurs des galaxies elliptiques, observées dans les amas les plus massifs de l’univers local. L’abondance des DSFGs à z > 4 se révèle donc être décisive pour contraindre ces modèles ainsi que pour vérifier notre compréhension globale de l’univers lointain. Le premier objectif de mon travail de recherche est d’identifier les candidates galaxies à z > 4 détectées sur des champs les plus larges possibles observés par Herschel et l’instrument SPIRE et d’examiner les propriétés statistiques de celles-ci. A cette fin, j’ai créé un nouvel algorithme de sélection dans le but d’augmenter substantiellement le nombre de candidates et de comprendre leur nature. / Over the last few decades, great progress has been made in our understanding of the star formation history of the Universe. With the discovery of distant, dusty star-forming galaxies (DSFGs) it has become apparent that observing at rest-frame UV and optical wavelengths is insufficient as a large fraction of the star formation is dust obscured. Thanks to the extensive observational studies carried out during the last two decades, we learn that DSFGs have a redshift peak at z ∼ 2, matching the cosmic time where galaxies have formed most of their young, massive stars. However, it remains extremely challenging to use the Herschel space observatory for identifying a tail extending towards much higher redshifts (z > 4). As a result, until recently only a small number of infrared-selected DSFGs at z > 4 were known, most of them strongly gravitationally lensed. One of the main goals of this Thesis is to assemble candidate z > 4 galaxies detected in a large area survey observed by Herschel-SPIRE and to examine the statistical properties and environments of these systems.