General relativity has a small number of known, exact solutions which model astronomically relevant systems. These models are highly idealized situations. Either perturbation theory or numerical simulations are typically needed to produce more realistic models. Numerical simulations are time-consuming and suffer from a difficulty in interpreting the results. In addition, global properties of numerical solutions are nearly impossible to uncover. On the other hand, standard perturbation methods are very difficult to implement beyond the second order, which means they barely scratch the surface of non-linear phenomena which distinguishes general relativity from Newtonian gravity. This work develops a method of finding exact solutions, inspired by perturbation theory, which have energy-momentum tensor components that approximately satisfy desired relationships. We find a spherical lump of matter which has a density profile $\mu \propto r^{-2}$ in a Robertson-Walker background; it looks like a galaxy in an expanding universe. We also find a plane-symmetric perturbation of a Bianchi type I metric with a density profile $\mu \propto z^{-2}$; it models a jet impacting a sheet-like structure. The former solution involves a wormhole while the latter involves a two dimensional singularity. These are both non-linear structures which perturbation theory can never produce.

Determining cosmological parameters from current observational data requires knowledge of the primordial density perturbations generated during inflation. We begin by examining a model of inflation along a flat direction of the minimal supersymmetric Standard Model (MSSM) and the power spectrum of perturbations it can produce. We consider the fine-tuning issues associated with this model and discuss a modification of the potential to include a hybrid transition that reduces the fine-tuning, without affecting the viability of the model. However, supersymmetric flat directions might play a role in other models of inflation as well. In particular, they may cause a feature in the primordial power spectrum of perturbations, unlike the scale-free spectrum assumed in the standard Lambda Cold Dark Matter (LCDM) cosmological model. We then show that in the presence of such a feature, an alternative cosmological model with a large local void and no dark energy provides a good fit to both Type Ia supernovae and the cosmic microwave background (CMB) data from the WMAP satellite. Constraints from the locally measured Hubble parameter, baryon acoustic oscillations and primordial nucleosynthesis are also satisfied. This degeneracy motivates a search for other independent observational tests of LCDM. The integrated Sachs-Wolfe (ISW) imprint of large-scale structure on the CMB is one such test. The ISW imprint of superstructures of size ~100 Mpc/h at redshift z~0.5 has been detected with >4 sigma significance, however it has been noted that the signal is much larger than expected. We revisit the calculation using linear theory predictions in a LCDM cosmology and find the theoretical prediction is inconsistent by >3 sigma with the observation. If the observed signal is indeed due to the ISW effect then huge, extremely underdense voids are far more common in the observed universe than predicted by LCDM.

The aim of this thesis is to build a paradigm to unravel the human mind’s fundamental kinship with the cosmos and its role as the vehicle of the universe’s unfolded meaning. The concept of ‘boundary’ is presented as a primary ontological force that drives, provokes and defines our thinking, consciously and subconsciously, in our attempt to achieve an understanding of self within the cosmos. It provides the hidden thread, the ‘limited concept’ that acts as a guide towards the building of this paradigm. Challenging its primarily physical interpretation, this thesis examines the concept of boundary from its genesis, imbedded in the primary moment of the birth of human consciousness within the universe, following it along its progressive complexity. Merging a primarily phenomenological with an epistemological approach by building on a number of essential evolutionary phases in our existence, through a synthesis of induction and deduction, we are confronted by how they are driven by boundary. Myth, religion, language, culture, philosophy, science, and even architecture are manifestations of humanity’s gradual attempt to understand, adapt to and transform our world and ourselves within it and in reference to it, displaying an inherent dynamic between our mind and our world. Bound in this dialectical creative opposition, our conceptualisations of the world are seen both as insights of our mind in its attempt to unravel the meaning of the cosmos, as well as the cosmos’s attempt to gradually reveal its nature within us, thus revealing their radical kinship. Ultimately, the aim is to reveal architecture and its embodied nature as a fundamental manifestation of our existence within the cosmos and to distil its message and purpose, its timeless task. Architecture is exposed as an existential medium, engaging the boundary between man and the cosmos, inviting us to read a model of the world while at the same time endowing us with our own sense of self and finally enabling us to coexist with our world in an interactive evolving equilibrium.

In this thesis I present the first gravitational lensing results from the Southern Cosmology Survey (SCS). I provide a preliminary study of an automated pipeline analysis of a large survey, in preparation for larger surveys. Future large-area sky surveys, such as Pan-STARRS-1 (PS1), have similar characteristics to the SCS data and will require full automation of the processing. Therefore, this data set provides an ideal test case to highlight the problems which will be faced by such surveys. To analyse the large SCS dataset, I develop an automated weak lensing pipeline based on the KSB. This pipeline has been rigorously verified using simulations and data which I detail here. Results are shown from a weak lensing analysis of 152 optically-selected clusters in 56 square degrees. I fit universal Navarro, Frenk and White (NFW) profiles to measure cluster masses, and use the relatively large area of the survey to test the universal shape of cluster profiles using stacking of the tangential shears. I present the first lensing mass measurements of Sunyaev-Zel’dovich (SZ) selected clusters. It has been long thought that SZ surveys would be a powerful way to detect galaxy clusters for cosmological studies. Simulations show that the SZ detection is independent of redshift and that the threshold corresponds very closely to a threshold in mass. It was, however, not guaranteed that the first blind SZ experiments would detect mass. Using optical imaging from the SCS, I present lensing masses for three clusters selected by their SZ emission in the South Pole Telescope survey (SPT). I confirm that the SZ selection procedure is successful in detecting mass concentrations and find that the SZ clusters have amongst the largest masses, as high as 15x1014M . Consequently I can confirm that the first installment of SZ detections has detected large mass concentrations. Using the best fit masses for all the clusters, I analytically calculate the expected SZ integrated Y parameter. Finally, the scaling relation of Reyes et al. (2008) of lensing Mlens 200 against optical L200 is tested over the redshift range z = 0:1 - 0:3 and extended to z = 0:3 - 0:8. While there is some discrepancy in the lower redshift-range, we agree with Reyes et al (2008) in the higherredshift sample if we assume no evolution of the scaling relation. To test the tangential shear profile of these clusters, 98 clusters are stacked. We find that by allowing the model to vary from an NFW, a very good fit can be found with a higher normalisation of the shears and a lower concentration. This study supports that of Mandelbaum et al. (2008) who show that that massive halos have a lower concentration than expected. Like the SCS, new large area surveys such as PS1 are not very deep, and it is crucial to understand not only how to analyse this size of dataset, but also the sort of results one could expect to achieve. I show in this thesis that 2D mass reconstructions can be done on data of this quality, and large galaxy clusters successfully reconstructed. With a number density of n ~ 9 it is possible to detect the most massive clusters with lensing, but it is difficult. With the lower number density of n ~ 6 or lower expected from PS1 it will prove very difficult to detect individual clusters. However, PS1 will survey a massive area, and so the stacking analysis should work extremely well, and it should be possible to further test the shape of the cluster profiles with stacking as I demonstrated here with the smaller SCS dataset.

Modern cosmology has reached an important juncture, at which the ability to make measurements of unprecedented accuracy has led to conclusions that are a fundamental challenge to natural science. The discovery that, in our current best model, the dynamics of the Universe are completely dominated by unseen dark matter and dark energy can do little but completely alter the shape of physics research in the 21st Century. Unfortunately,much of our insight into these phenomenamust come from observations of visible matter alone; this raises serious problems, as the tracing of dark matter by visible matter is as yet poorly understood. Gravitational lensing offers strong prospects for probing the interwoven history of dark and visible matter, as mass in any form may be detected where it exists untraced by baryons. In this Thesis I describe advances made in the field of weak gravitational lensing, which constrains the properties of the matter distribution on cosmological scales using a statistical analysis of the coherent gravitational distortions of distant galaxy images. I summarize the development of gravitational flexion, a higher order extension to traditional weak lensing, and describe my work done to bring the study of flexion to a stage where it may be employed to make accurate cosmological measurements. I show how flexion is sensitive to matter structure on smaller physical scales than existing lensing techniques and, therefore, promises to shed new light upon key untested predictions of cosmological models if it can be measured to sufficient accuracy. I discuss the success of my efforts in this direction, and describe the issues to be encountered in the careful analysis of this subtle gravitational signal. This research has involved advances in many areas: the calculation of theoretical flexion predictions, the refinement of image analysis methods for accurate galaxy shape estimation, and the practical application of these new flexion techniques to extragalactic imaging data. The culmination of these efforts is a new maximum likelihood analysis of the galaxy-galaxy lensing signal in the Hubble Space Telescope Galaxy Evolution from Morphology and SEDs (GEMS) Survey, incorporating improvements and modifications necessary for the combination of flexion with traditional weak lensing measurements. The results of this work, and particularly the extent to which measurements of flexion provide extra cosmological insight, are discussed in detail. The conclusion is a summary of all that has been learned about the use of flexion as an accurate probe of cosmology, and a discussion of its prospects for answering some of the many questions that remain about dark matter. Within the next few year wide-area survey telescopes will begin imaging huge volumes of deep space, with the measurement of the gravitational lensing signal being given high priority in the analysis of these data. Within this context, the primary inquiry of this Thesis is the extent to which the application of flexion measurement techniques will help shed new light upon the unseen, and currently poorly understood, components of the Universe.

Searching for active galactic nuclei (AGNs) in dwarf galaxies is important for our understanding of the seed black holes that formed in the early universe. Here, we test infrared selection methods for AGN activity at low galaxy masses. Our parent sample consists of similar to 18,000 nearby dwarf galaxies (M-* < 3 x 10(9) M-circle dot, z < 0.055) in the Sloan Digital Sky Survey with significant detections in the first three bands of the AllWISE data release from the Wide-field Infrared Survey Explorer (WISE). First, we demonstrate that the majority of optically selected AGNs in dwarf galaxies are not selected as AGNs using WISE infrared color diagnostics and that the infrared emission is dominated by the host galaxies. We then investigate the infrared properties of optically selected star-forming dwarf galaxies, finding that the galaxies with the reddest infrared colors are the most compact, with blue optical colors, young stellar ages, and large specific star formation rates. These results indicate that great care must be taken when selecting AGNs in dwarf galaxies using infrared colors, as star-forming dwarf galaxies are capable of heating dust in such a way that mimics the infrared colors of more luminous AGNs. In particular, a simple W1 - W2 color cut alone should not be used to select AGNs in dwarf galaxies. With these complications in mind, we present a sample of 41 dwarf galaxies that fall in the. WISE infrared color space typically occupied by more luminous AGNs and that are worthy of follow-up observations.

Many physical theories beyond the Standard Model predict time variations of basic physics parameters. Direct measurement of the time variations of these parameters is very difficult or impossible to achieve. By contrast, measurements of fundamental constants are relatively easy to achieve, both in the laboratory and by astronomical spectra of atoms and molecules in the early universe. In this work, measurements of the proton to electron mass ratio mu and the fine structure constant alpha are combined to place mildly model-dependent limits on the fractional variation of the quantum chromodynamic scale and the sum of the fractional variations of the Higgs vacuum expectation value (VEV) and the Yukawa couplings on time-scales of more than half the age of the universe. The addition of another model parameter allows the fractional variation of the Higgs VEV and the Yukawa couplings to be computed separately. Limits on their variation are found at the level of less than 5 x 10(-5) over the past 7 Gyr. A model-dependent relation between the expected fractional variation of a relative to mu tightens the limits to 10(-7) over the same time span. Limits on the present day rate of change of the constants and parameters are then calculated using slow roll quintessence. A primary result of this work is that studies of the dimensionless fundamental constants such as a and mu, whose values depend on the values of the physics parameters, are excellent monitors of the limits on the time variation of these parameters.

The potential of using cluster clustering for calibrating the mass-richness relation of galaxy clusters has been recognized theoretically for over a decade. Here, we demonstrate the feasibility of this technique to achieve high-precision mass calibration using redMaPPer clusters in the Sloan Digital Sky Survey North Galactic Cap. By including cross-correlations between several richness bins in our analysis, we significantly improve the statistical precision of our mass constraints. The amplitude of the mass-richness relation is constrained to 7 per cent statistical precision by our analysis. However, the error budget is systematics dominated, reaching a 19 per cent total error that is dominated by theoretical uncertainty in the bias-mass relation for dark matter haloes. We confirm the result from Miyatake et al. that the clustering amplitude of redMaPPer clusters depends on galaxy concentration as defined therein, and we provide additional evidence that this dependence cannot be sourced by mass dependences: some other effect must account for the observed variation in clustering amplitude with galaxy concentration. Assuming that the observed dependence of redMaPPer clustering on galaxy concentration is a form of assembly bias, we find that such effects introduce a systematic error on the amplitude of the mass-richness relation that is comparable to the error bar from statistical noise. The results presented here demonstrate the power of cluster clustering for mass calibration and cosmology provided the current theoretical systematics can be ameliorated.

We present the initial results from a program aimed at establishing a network of hot DA white dwarfs to serve as spectrophotometric standards for present and future wide-field surveys. These stars span the equatorial zone and are faint enough to be conveniently observed throughout the year with large-aperture telescopes. The spectra of these white dwarfs are analyzed in order to generate a non-local-thermodynamic-equilibrium model atmosphere normalized to Hubble Space Telescope colors, including adjustments for wavelength-dependent interstellar extinction. Once established, this standard star network will serve ground-based observatories in both hemispheres as well as space-based instrumentation from the UV to the near IR. We demonstrate the effectiveness of this concept and show how two different approaches to the problem using somewhat different assumptions produce equivalent results. We discuss the lessons learned and the resulting corrective actions applied to our program.

This thesis presents results from two simulation studies of galaxy formation. In the first project, a dark-matter-only HORIZON simulation is used to investigate the environment and redshift dependence of mass accretion onto haloes and subhaloes. It is found that the halo accretion rate varies less strongly with redshift than predicted by the Extended Press--Schechter formalism, and that low accretion events may drive the radio-mode feedback hypothesized for recent galaxy formation models. The subhaloes at $z<0.5$ in the simulation accrete at higher rates than haloes, on average, and it is argued that this is due to their enhanced clustering at small scales. There is no dependence of accretion rate on environment at $zsim2$, but a weak correlation emerges at $zleq0.5$. The results further support previous suggestions that at $z>1$, dark matter haloes and their associated black holes grew coevally, but imply that haloes could be accreting at fractional rates that are up to a factor of 3--4 higher than their associated black holes by the present day. In the second project, outputs from one of the Adaptive Mesh Refinement NUT simulations are analyzed in order to test whether filamentary flows of cold gas are responsible for the build-up of angular momentum within a Milky Way type disk at $zgeq3$. A set of algorithms are presented that use the resolved physical scale of $12,mathrm{pc}$ to identify: (i) the central gas disk and its plane of orientation; (ii) the complex individual filament trajectories that connect to the disk, and; (iii) the infalling satellites. The results suggest that two filaments at $zgtrsim 5.5$, which later merge to form a single filament at $zlesssim 4$, drive the angular momentum and mass budget of the disk between $3lesssim zlesssim 8$, whereas luminous satellite mergers make negligible fractional contributions. These findings hence provide strong quantitative evidence that the growth of thin disks in low mass haloes at high redshift is supported via inflowing streams of cold gas.