The technology described here lends itself particularly well to the detection of the <i>B-</i>mode polarisation signal intrinsic to the Cosmic Microwave Background Radiation (CMBR). It is unclear as to whether the <i>B</i>-mode polarisation detection limit will be due to detector sensitivity or due to foreground contamination. Part 1 focuses on fabricating the best possible devices, where sensitivity approaches the phonon noise limit, whereas in Part 2 I predict a limit presented by extragalactic foreground radio sources at frequencies approaching the CMBR observing frequency. The planar technology presented in Part 1 consists of a microstrip-coupled transition edge sensor (TES) that uses a finline taper to transfer between waveguide and superconducting Nb microstrip. The microstrip transports the signal to a matched Au/Cu resistor, which is deposited on a thermally isolated SiN membrane. The dissipated RF power causes the resistance of a Mo/Cu TES bilayer to increase, and the resulting reduction in bias current is read out by a SQUID. This technology may be combined with waveguide-horn technology to produce sensitive bolometric detectors with well-defined, single-mode beam patterns and polarisation characteristics. Finally, in Part 2 I concentrate on 43 sources from the 9C follow up survey. The data is used to provide insight into the population of radio foreground sources at 5 and 22 GHz in Stokes parameters <i>I, Q </i>and <i>U. </i>These values are then used to predict the Angular Power Spectrum (APS) contribution of the sample, which are compared to the APS of the <i>B</i>-mode polarisation signal that is intrinsic to the CMBR. This part of the Thesis contributes to the small amount of published work on the feasibility of detecting the <i>B</i>-mode polarisation signal of the CMBR by considering the foreground contribution of extragalactic radio sources.

This thesis considers if the current 'unified scheme' for powerful extragalactic radio sources are valid in the light of cosmological tests embodied in radio source count and identification data. Although the analysis is based on the simplest possible parametric descriptions of radio source evolution and beaming, the overall agreement with the available observational data is quite remarkable. The first part of the thesis describes an identification programme of flat-spectrum radio sources at 2.7 GHz. The complete sample provides two important pieces of information (1) whether a turnover in the space density of quasars as a function of redshift can be determined (2) the nature of the flat-spectrum galaxies in the sample which must be understood in the context of the current unified scheme. A statistical analysis of the space-density evolution of the powerful radio sources is then performed, adopting simple parametric descriptions of the evolution to fit a low-frequency radio source count. The analysis uses a coherent description of the known radio source populations such that the space density evolution of the powerful radio sources can be established. The successful space density models are scrutinized to ensure that these unified scheme-based predictions are reconcilable with all available data, which includes the observed turnover in space density of the FRII population from the 2.7-GHz flat-spectrum sample. The successful space density evolution model is then applied to the 5-GHz source count where the contribution from Doppler-beamed, flat-spectrum, sources is significant. The unified scheme posits that the beamed products have the same evolution histories as their underlying 'parent' sources, so that the space density model is valid at all radio frequencies. Simple parametric descriptions of the Doppler beaming are applied to the powerful radio source populations to produce their respective flat-spectrum products.

This thesis presents the detection of weak gravitational lensing by large-scale structure, also known as 'cosmic shear'. This effect, in which the images of faint background galaxies are coherently distorted by the gravitational potential of intervening foreground structures, is poised to become the most direct measure of the mass fluctuations in the universe. It is sensitive to all components of mass present along a line of sight, whether visible matter or dark matter, and therefore does not suffer from poorly-justified assumptions in relating the two. In order to make a pioneering detection of cosmic shear, a strategy for observing the effect is developed here in the context of deep-field observations with the 4.2m William Herschel Telescope (WHT). The criteria for detecting the cosmic shear signal, and the resulting survey, are described in detail. The procedure for image data reduction is then explained. As the distortions expected from cosmic shear are very small, we require a method for measuring the shapes of galaxies to great precision, and a painstaking treatment of systematic effects. Various approaches to shape measurement are discussed, and the adopted approach using the Kaiser, Squires and Broadhurst (KSB) method is described. This is followed by a study of this algorithm's performance upon artificial sheared images, produced using detailed realisations of a galaxy ensemble drawn from the Hubble Space Telescope Groth strip, and including realistic systematics reproducing observational conditions for the WHT. It is demonstrated that, despite low-level biases in KSB, the method is capable of measuring cosmic shear signals. The secure detection of cosmic shear on the observed WHT fields is then described. The analysis and removal of systematic effects due to telescope optics and tracking are discussed, and the shear estimators to be used are considered in detail. The significance of detection is calculated, and the cosmological implications of this detection are discussed. It is found that our measurements are consistent with the predictions of cluster-normalised Cold Dark Matter (CDM) models, but a COBE-normalised Standard CDM model is ruled out at high significance. For the currently favoured cosmological constant-dominated ΛCDM model, our measurement provides a constraint on the normalisation of the mass power spectrum fully consistent with that derived from cluster abundances. The potential effect of intrinsic alignments between galaxies is considered. Following the detection of cosmic shear, precision measurements of the effect become possible, together with initial measurements of the amplitude of the mass power spectrum. For this purpose, a large survey with the 10m Keck II telescope is described, along with an extended WHT data set. The use of correlation functions is explained, and current constraints on the fluctuations of matter and the density of the Universe from these weak lensing surveys are given.

The X-ray background (XRB) and the intergalactic medium (IGM) are two of the most interesting topics in X-ray cosmology. Historically, studies of either subject provided important information on the other. For more than 30 years, the problem of the origin of the XRB has remained unsolved, despite large advances in X-ray instrumentation. Along the way, important constraints have been placed on the state of the intergalactic medium. Based on theoretical modelling and observational constraints, I have investigated both the origin of the X-ray background and the physical state of the intergalactic medium in this <I>thesis</I>. The first Chapter gives an outline of the observed properties of the XRB and the manner by which X-ray studies constrain the IGM. Chapter 2 reviews previous observations of the XRB and our current understanding of the spectrum, isotropy, and the resolved component of the XRB. An analysis of ROSAT data to study the spatial structure of the XRB is presented in Chapter 3. The ROSAT PSPC data include one deep survey field and six colinear medium survey fields, from which I have obtained limits on excess fluctuations and the autocorrelation function (ACF) of the XRB in the soft X-ray energy band from 1 arcmin to 5 deg. The 2σ upper limits of these results show no strong signal of anisotropy and inhomogeneity. Incorporating the observed ACF into detailed models of evolving discrete sources, I find that the correlation length of the major source dominating the residual XRB has to be < 4<I>h</I><SUB>50</SUB><SUP>-1</SUP> Mpc, which is less than that of normal galaxies. Spectral fitting of both resolved and unresolved components of the XRB was carried out and is shown in Chapter 4. Data from the QSF3 field observed with the ASCA SIS and the ROSAT PSPC were used for this task. The diffuse XRB has a flat spectrum, which is well fit by a single power-law extending from 0.1 to 7 keV, with a photon index ranging from 1.4 to 1.5. On the contrary, the spectrum of the resolved component appears steeper. Such a steep-spectrum component can contribute in total less than 30 per cent to the residual XRB below 3 keV and only a few per cent above 3 keV. This suggests the entire XRB is composed of two spectrally distinct populations, of which the one with a steeper spectrum has been resolved in the soft band and most of which are AGN.

A rapid binary evolution algorithm is produced that enables modelling of even the most complex binary systems. In addition to all aspects of the single star algorithm, features such as mass transfer, mass accretion, common-envelope evolution, collisions, supernova kicks and angular momentum loss mechanisms are included. Circularization and synchronization of the orbit owing to tidal interactions are calculated for convective, radiative and degenerate damping mechanisms. The algorithm is used to study the formation and evolution of various binary systems. The effect that tidal friction has on the outcome of binary evolution is investigated. The rapid binary code is used to generate a series of large binary populations enabling the formation rate of interesting individual species and events to be calculated. By comparing the results for populations with and without tidal friction it is possible to quantify the hitherto ignored systematic effects of tides and how that modelling of tidal evolution in binary systems is necessary in order to draw accurate conclusions from population synthesis work. The rapid single star and binary star evolution algorithms are incorporated into a state-of-the-art <I>N</I>-body. This code is used to explore the evolution of a primordial binary population as a function of radius in a globular cluster. HST images of the young rich LMC cluster NGC 1818 have revealed a binary fraction that increases towards the cluster centre, from ~ 20 ±5% in the outer parts, to ~ 35 ± 5% inside the core. A comparison with the results of the <I>N</I>-body models suggests that this is consistent with dynamical mass segregation and need not be primordial. In order to investigate the incidence and distribution of blue straggler starts in a cluster environment, <I>N</I>-body models are generated for comparison with observations of the old open cluster M67. The effectiveness of proposed blue straggler formation mechanisms, such as mass transfer within a binary system or the collision of two main sequence stars, is tested. The nature and extent of the binary population required to produce the observed numbers of blue stragglers is constrained. The escape of stars from the tidal radius of a cluster as it evolves is discussed and an estimate of the escape rate given.

The data from Cosmic Microwave Background (CMB) experiments are becoming more complex with each new experiment. A consistent way of analysing these data sets is required so that direct comparison is possible between the various experimental results. This thesis presents several techniques that can be used to analyse CMB data. A summary of the processes involved in the formation of the fluctuations and the resultant radiation we expect to see in our sky maps is presented. This includes both the CMB and other emission components that operate at the frequencies covered by CMB experiments. These other components are the Sunyaev-Zel'dovich effect, point source emissions and various Galactic foregrounds (namely dust, bremsstrahlung and synchrotron emission). Various present and future experiments are discussed. Preliminary analysis is done on existing data taken from the Tenerife and Jodrell Bank experiments. This data is used to put constraints on Galactic emissions and point source contributions. The data discussed covers the 5 GHz to 33 GHz frequency range which is at the lower end of the useful CMB spectral window. Lower frequency surveys are used to put constraints on the spectral index of some of the Galactic foregrounds (the frequency range is not large enough to put constraints on dust emission which is expected to dominate at higher frequencies). The Tenerife experiments are used to put constraints on the level of the CMB anisotropy. Maximum Entropy Methods (MEM), the Wiener filter, CLEAN and Singular Value Decomposition are introduced. These four methods offer different alternatives to find the best underlying cosmological signal within the noisy data. The usual approach of a positive-only Maximum entropy is enlarged to cover both positive and negative fluctuations, as well as multifrequency and multicomponent observations. Simulations performed here show that MEM is the best method of the four tested when attempting a reconstruction of the CMB signal. The final sky maps of the CMB produced with the Maximum Entropy algorithm, as well as maps of the various contaminants that the experiments are also sensitivity to, are presented. The maps from a range of different experiments can be used to put constraints on various cosmological parameters such as the density parameter, Ω<SUB>o</SUB>, Hubble's constant, <I>H</I><SUB>o</SUB>, and the spectral index of the large scale CMB power spectrum, <I>n</I>. Subsequent analysis performed on the sky maps is presented. These include examining the topology using genus as well as looking at the power spectrum and correlation functions. The methods discussed are first applied to simulations to test their usefulness at distinguishing between the origins of the fluctuations and then applied to the reconstructed CMB sky maps. New constraints on the power spectrum and some of the cosmological parameters will be given in the final chapter. Here, the data and analysis methods described will be summarised and the future of CMB experiments discussed.

In this thesis, we study the effect of feedback on the formation and evolution of galaxies using a wide range of techniques, including observations, semi-analytic modelling and numerical simulations. There is now a good understanding of the gravitational evolution and clustering of dark matter haloes. In contrast the study of feedback processes, that regulate the formation and growth of the baryonic component in galaxies is still in its infancy. In this thesis, feedback is defined to be a process involving the baryonic component that invokes a back reaction of the process on itself or on the causes that produced it and examples include feedback from supernova and AGNs, radiative feedback and feedback induced by shocks. From an observational near-infrared survey we extract a sample of old elliptical Extremely Red Objects (EROs) at <i>z</i> ~ 1. We find a relatively high number density of massive galaxies at <i>z</i> ~ 1, in agreement with recent observations of cosmic downsizing. The observed space density of old elliptical galaxies is compared to predictions from galaxy formation models. We find that standard hierarchical models underpredict the number of old EROs at <i>z</i> ~ 1, suggesting that additional feedback processes not included in the models, such as AGN feedback might be required to reproduce the observations. Using our ERO sample we study a population of sources with extremely high mid-to-near infrared flux ratios and show that this sample broad band selection method selects primarily sources with extremely obscured AGN activity at <i>z</i> ≳ 1. Our results indicate that obscured AGN activity contribute ~ 20% of the observed cosmic IR background and that a significant fraction of obscured AGN activity is still missed by X-ray surveys. We have developed an analytic model of supernova feedback to study the fraction and metallicity of ejected gas, the relationship between outflow and star formation rates and the mass-metallicity relation of a galaxy population as a function of redshift. We find that our physically motivated, though simplified model of the interstellar medium and star formation produces results that are in good agreement with a range of observations. The model is applied to calculate the metallicity evolution and statistical properties of damped Lyman-α systems (DLAs). The combined effects of star formation, outflows and infall in our model produces a mild evolution in the <i>N</i>_HI-weighed metallicity and a nearby constant evolution of Ω_HIwith redshift in the range of <i>z</i> = 0 – 5, consistent with observations.

No description available.

This thesis concerns the behaviour of eccentricity in accretion discs in binary systems, both stellar binaries and systems with a single planet. Linear equations are derived which describe the precision of eccentric discs and growth of eccentricity in two-dimensional and three-dimensional discs, with adiabatic or locally isothermal thermodynamics. The gravitational interaction of the binary companion with the disc and of the disc particles with each other is also included in the equation. These equations are analysed by looking for normal modes, with eigenvalues which contain the precession and growth rates for eccentricity. It is shown that the eigenvalue can be written in terms of integrals which describe the separate contributions of the different fluid and dynamical terms to precession and growth. It is also described how simple forms of the eccentricity equation can be interpreted by rewriting them as time-dependent Schrödinger equations. They are applied to explain the behaviour of real systems, firstly superhump systems in which a resonant tidal instability is thought to cause strong outbursts. It is shown that the presence of a strong delta-function resonance creates a cusp in the eccentricity distribution and results in suppression of eccentricity at the resonance leading to low growth rates or stability. A more careful treatment of the resonance results in the resonance causing precession of the disc as well as eccentricity growth and that this leads to a strong eccentric instability in the disc. When the calculated precession rates are compared with observation it is found that a simple 2D disc without a resonance provides the best fit for the expected system parameters. The eccentricity equations are then applied to the interaction of a single planet, large enough to clear a gap, with a protoplanetary disc. It is shown that the width of the gap cleared by a planet is the critical parameter for determining whether the eccentricity can grow, and that a wider gap is needed to excite the planet’s eccentricity than the disc’s. Both eccentric Lindblad and eccentric corotation resonances need to be considered to explain eccentricity growth. This may lead to a new model for the formation of eccentrically orbiting planets in which the disc’s eccentricity is excited first and feedback processes than cause the planet to become eccentric.

The halo of the Milky Way galaxy hosts a substantial quantity of neutral and ionised gas in the form of complexes with large angular extents. Many of these have anomalously large velocities that cannot be explained in simple models of differential Galactic rotation. These so-called high-velocity clouds, some of which are elongated stream-like structures, have been an enigmatic component of the Galactic halo since their discovery in the mid 20^th century. The absence of stars within these objects makes it difficult to determine their distance, without which other physical properties, such as size and mass, cannot be constrained. Our general lack of understanding of their origin stems mainly from these uncertainties. In this thesis, a number of so-called geometrodynamical methods are developed in order to constrain the distances to high-velocity HI structures in the Galactic halo. In the first approach, I develop a method that leads to the successful calculation of the distance to the Magellanic stream, without prior knowledge of where the progenitors (the Large and Small Magellanic Clouds) of the stream lie. This approach is restricted to structures with a substantial elongation, a condition not satisfied by other complexes of high-velocity clouds. In Chapter 3, I therefore investigate how the possibility of associations between different, less extended halo structures may lead to the determination of their distances. For this study, I begin by presenting a technique that enables transverse velocities to be calculated for structures that exhibit at least some degree of elongation, and for which sky positions and line-of-sight velocities are known. Using this velocity information, possible orbits can then be deduced in a distance-dependent manner. The validity of the method is verified through its successful application to the stellar Sagittarius stream. I also find two sets of candidates for possible associations, one between the Sagittarius stream and a high-velocity cloud, and the other between two high-velocity clouds. Chapter 4 then explores the degree to which analytic forms can accurately reproduce orbits within a power-law potential. In the course of these three chapters, I have also investigated the orbital characteristics and dynamics of objects in differing potentials. The penultimate chapter of this thesis describes a different problem, where I investigate the possibility of massive black holes as the dark matter component of dwarf spheroidal galaxies. I find that the heating of the stellar population through two-body encounters with the black holes leads to its rapid expansion. This scenario therefore requires the dwarf galaxy to have been significantly more compact in the past, in which case the mass of the individual black holes is constrained to be less than a few 10⁵_M⊙.