School of Physics - Theses
Permanent URI for this collection
Search Results
1 - 5 of 5
-
ItemObservational methods towards constraining the chemical evolution of galaxies( 2020)Understanding the array of physical processes that have shaped galaxy assembly remains one of the most fundamental pursuits in astrophysics. Gas in galaxies is enriched with heavy elements via stellar nucleosynthesis, but chemical abundances (``metallicity'') are also shaped by galaxy-scale processes including gas accretion, feedback-driven outflows, radial gas flows, interactions, and mergers. Metallicity measurements therefore afford one of our most powerful observational probes of galaxy evolution. In this thesis I explore the performance of observational methods for constraining (i) gas-phase metallicity in galaxies, and (ii) host dark matter halo masses of galaxies; the latter of which is critical to the physics of gas flows due to its contribution to the gravitational potential well of galaxies. A particular focus is the improved understanding of systematic uncertainties near instrumental limits, which will be vital to maximise the impact of surveys conducted with future facilities. Galaxy clustering is an efficient approach for drawing statistical connections between galaxies and their host dark matter haloes, however traditional methods are challenging to apply at z > 2 where imaging survey volumes are limited. I instead apply a counts-in-cell approach to photometric z ~ 2 candidates from a random-pointing Hubble Space Telescope survey, showing mean counts of N > ~5 per field are capable of constraining the large scale galaxy bias. The James Webb Space Telescope will achieve comparable number counts out to z ~ 8, and thus a similar JWST survey could place novel constraints on the halo masses of galaxies in the epoch of reionization. Global metallicities in low-mass galaxies afford important constraints on the impact of feedback-driven outflows on galaxy evolution. However at high-z, obtaining the requisite emission line measurements is observationally challenging. I use Keck/MOSFIRE spectroscopy to explore prospects for extending z ~ 1 - 2 metallicity measurements to lower masses. I find the dominant source of uncertainty arises from reduced number of emission lines as opposed to lower signal-to-noise, even at the detection limit. JWST/NIRSpec will revolutionise high-z metallicity studies due to the large suites of emission lines it will be able to assemble. Electron temperatures (T_e) measured with auroral lines are an important baseline in metallicity studies. However the faintness of auroral lines has hitherto limited spatially resolved T_e studies. I report two separate studies based on mapping auroral lines in integral-field spectroscopy (IFS) of low-z galaxies. Measurements of auroral lines in the SAMI Galaxy Survey afford new insights into the effects of ionisation parameter variations on recovered metallicity gradients. Applying these principles to Keck/KCWI IFS data of an edge-on disk galaxy, I measure an extra-planar temperature gradient and present preliminary evidence for extra-planar metallicity variations.
-
ItemThe host galaxies of high-redshift quasars( 2020)In the early Universe, we observe supermassive black holes with masses of up to a billion times the mass of the Sun, accreting at or even above the Eddington limit. These high-redshift quasars are some of the most luminous objects in the Universe, and raise many questions about the formation and growth of the first black holes. Investigating their host galaxies provides a useful probe for understanding these high-redshift quasars. In the local Universe, there are clear correlations between the mass of a supermassive black hole and the properties of its host galaxy, indicating a black hole--galaxy co-evolution. Exploring how these black hole--host relations evolve with redshift can give valuable insights into why these relations exist. Studying the host galaxies of high-redshift quasars thus provides vital insights into the early growth of supermassive black holes and the black hole--galaxy connection. In this thesis I use three techniques to study the host galaxies of high-redshift quasars: the Meraxes semi-analytic model, the BlueTides hydrodynamical simulation, and observations with the Hubble Space Telescope. Meraxes is a semi-analytic model designed to study galaxy formation and evolution at high redshift. Using this model, I study the sizes, angular momenta and morphologies of high-redshift galaxies. I also use Meraxes to study the evolution of black holes and their host galaxies from high redshift to the present day. The model predicts no significant evolution in the black hole--host mass relations out to high redshift, with the growth of galaxies and black holes tightly related even in the early Universe. I also examine the growth mechanisms of black holes in Meraxes, finding that the majority of black hole growth is caused by internal disc instabilities, and not by galaxy mergers. I then use the BlueTides cosmological hydrodynamical simulation to investigate the detailed properties of quasar host galaxies at z=7. I find that the hosts of quasars are generally highly star-forming and bulge dominated, and are significantly more compact than the typical high-redshift galaxy. Using BlueTides I make predictions for observations of quasars with the James Webb Space Telescope, finding that detecting quasar hosts at these redshifts may be possible, but will still be challenging with this groundbreaking instrument. Finally, I use observations from the Hubble Space Telescope to obtain deep upper limits on the rest-frame ultraviolet luminosities of six z~6 quasars. I also detect up to 9 potential companion galaxies surrounding these quasars, which may be interacting with their host galaxies. Observations with the upcoming James Webb Space Telescope are needed to detect quasar host galaxies in the rest-frame ultraviolet and optical for the first time.
-
ItemSimulations of source recovery and completeness in galaxy surveys at high redshift( 2018)The search for and characterisation of galaxies at high-redshift is a very active topic in Astrophysics. Thanks to advances in observations from space, the redshift frontier is approaching the epoch of formation of first generation objects. Thus, these samples of galaxies can give us insight into the processes that govern galaxy formation and evolution. One of the key observables used to characterise galaxy populations throughout the cosmic history is their luminosity function (number of galaxies per unit luminosity per unit volume), which requires knowledge and characterisation of the completeness and selection functions of a survey, in addition to the catalogue of discovered objects. In this thesis, we present a search for high-redshift galaxies (redshift z > 6) in two in the Hubble Space Telescope surveys, the Brightest of Reionizing Galaxies Survey (BoRG), and the Reionization Lensing Cluster Survey (RELICS) using a photometric selection technique (the Lyman break dropout selection). We aim at using the resulting galaxy candidates to estimate a new measurement of the luminosity function at z ~ 10. To achieve that, we develop GLACiAR, an open Python-based tool available on GitHub, which is designed to estimate the completeness and selection functions in galaxy surveys. The code is tailored for multiband imaging datasets aimed at searching for high-redshift galaxies through the Lyman Break technique, but it can be applied broadly. The code generates artificial galaxies that follow Sérsic profiles with different indexes and with customisable size, redshift and spectral energy distribution properties, adds them to input images, and measures the recovery rate. We finally apply GLACiAR to quantify the completeness and redshift selection functions for J-dropouts sources (redshift z ~ 10 galaxies). Our comparison with a previous completeness analysis on the same dataset shows overall agreement, but also highlights how different modelling assumptions for artificial sources can impact completeness estimates.
-
ItemThe fundamental plane and peculiar velocities from the 6dF galaxy survey( 2012)Early-type galaxies (ellipticals and lenticulars) are observed to populate the relation known as the Fundamental Plane that links their effective radius, R_e, stellar velocity dispersion, σ, and mean surface brightness, I_e. We have measured Fundamental Plane parameters in the near-infrared J, H and K passbands for ~10^4 of the brightest early-type galaxies in the 6dF Galaxy Survey (6dFGS). We improve upon previous regression techniques used to derive the Fundamental Plane by developing a robust maximum likelihood algorithm for fitting the galaxy distribution in Fundamental Plane space with a 3D Gaussian model. We exploit this large near-infrared-selected sample of galaxies to investigate trends in the Fundamental Plane with stellar population, morphology and environment. The 6dFGS galaxies exhibit clear stellar population trends in Fundamental Plane space, with age varying most strongly orthogonal to the plane. Remarkably, none of the stellar population parameters vary along the long axis of the plane, which corresponds to luminosity density. The Fundamental Plane slopes show little variation with either morphology or environment, but the Fundamental Plane size zeropoint is systematically larger for galaxies in lower density environments and for early-type spiral bulges. We speculate that age drives all the trends with residuals about the plane through its correlation with environment, morphology and metallicity. Using the Fundamental Plane, we measure distances and peculiar velocities for ~10^4 6dFGS galaxies to form the largest and most homogeneous peculiar velocity sample to date. Using a maximum-likelihood approach, we measure the overall bulk galaxy motions from the 6dFGS velocity field for the local volume of the universe, finding broad agreement with the predicted velocity field constructed from the 2MASS Redshift Survey. The local volume out to 16 120 km/s is found to have a bulk motion of 337 km/s in the direction (l,b) = (313°±9°,14°±10°), in good agreement with the results of other recent studies. A comparison of the observed and predicted fields is used to constrain parameters relating the distribution of galaxies and matter. We obtain a redshift-space distortion parameter β = 0.29±0.06 and a bias parameter for the 6dFGS velocity sample of b = 1.69±0.36. The 6dFGS velocity field provides an independent probe of cosmological parameters defining models of large-scale structure formation. Next steps include: (i) combining the 6dFGS sample in the south with the SDSS sample in the north to obtain an all-sky velocity field; (ii) deriving additional constraints on cosmological parameters from the velocity power spectrum analysis; and (iii) comparing the Fundamental Plane distances and velocities for early-type galaxies with the Tully-Fisher distances and velocities for spiral galaxies that will be obtained with the WALLABY survey on the Australian SKA Pathfinder.
-
ItemThe broad emission line region of quasars and gravitational lensing by early-type galaxies( 2012)This thesis has focused on predicting emission line flux ratios from the broad emission line region of quasars under different physical conditions, and measuring the dark matter fraction and total mass density slope within early-type galaxies using gravitational lensing. Quasars are the energetic cores of distant galaxies, and they reside in some of the oldest, most massive objects formed in the universe. Due to their incredible luminosity (as much as $10^5$ times greater than a typical galaxy), quasars can be observed at extremely large distances. Quasars have a unique spectrum, with bright, broad emission lines that are produced by photoionised gas that is close to the central super-massive black hole. Despite the prominence of these broad emission features, the gas physical conditions and the geometry of the emission region are poorly understood. Due to its small scale and large distance, the emission line region cannot be resolved directly — even with the most powerful telescopes — and simulations are required to understand the mechanism that produces the unique quasar spectrum. Using simulations of micro-physical processes, including photoionisation, the broad emission line flux ratios can be calculated for a range of gas densities and distances from the central black hole. Using the photoionisation code, Cloudy, hydrogen and helium line emission was over the range of possible broad emission line region conditions. The hydrogen and helium lines are of particular interest because the line emission has strong dependence on the gas number density and incident ionising flux, whilst having only a negligible dependence on several other free parameters of the model. These simulations were then used to find a set of interesting ratios that can be used to determine the limits on the upper limit on the gas number density, and outer radius of the emission region. This thesis demonstrates a new technique for determining the physical conditions of the broad line emitting gas in quasars, using optical and near-infrared hydrogen and helium emission lines. Near-infrared line ratios are advantageous, as they have a negligible dependence on the amount of internal dust. A locally optimally emitting cloud model of the broad emission line region was applied to four nearby (z $\sim$ 0.2) quasars from the Glikman et al. (2006) sample. By comparing simulated emission line ratios to measured ratios from optical and near-infrared spectroscopy, the physical conditions required to produce the observed emission lines were inferred. The model provides a good fit to three of the objects, and a fair fit to the fourth object. We find that low incident ionising fluxes ($phi <10^{18}$cm^-2 s^-1), and high gas densities (n>10^{12} cm^-3) are required to reproduce the observed line ratios. This analysis demonstrates that the use of composite spectra in photoionisation modelling is inappropriate; models must be fitted to the individual spectra of quasars. This thesis also derives properties of early-type galaxies using a joint gravitational lensing and stellar-dynamics analysis. The sample consists of 11 early-type galaxies from the Strong Lenses in the Legacy Survey (SL2S). The median deflector redshift is 0.5, making it the largest sample of intermediate redshift lenses that have been studied using a joint lensing and dynamics analysis. By combining measured redshifts and stellar velocity dispersions from Keck spectroscopy with lens models from Gavazzi et al. (2012, submitted), the total mass density slope inside the Einstein radius for each of the 11 lenses was derived. The average total density slope was found to be 2.16$\pm$0.9, with an intrinsic scatter of 0.25. The dark matter fraction for each lens within half the effective radius was also determined. The average projected dark matter mass fraction was found to be 0.42$\pm$0.08 with a scatter of 0.25 for a Salpeter initial mass function. By combining the SL2S results with those from previous studies, a mild trend in the cosmic evolution of the total mass density slope was found. This suggests that the total density profile of massive galaxies has become slightly steeper over cosmic time. If this result is confirmed by larger samples, it would indicate that either dissipative processes or off-axis major mergers play an important role in the growth of massive galaxies since a redshift of 1.