Bibliography: leaves 76-81. / The intention of this project was primarily to determine the distance to the Galactic Centre using a new technique pioneered by Paczynski and Stanek (1998) that uses the red clump stars as a standard candle. This was made possible by the large number of stars simultaneously observed by the OGLE microlensing project which yields well-defined red clump structures in their colour magnitude diagrams. We have used data obtained at Sutherland in July 1997, in collaboration with the PLANET microlensing project. It was hoped that the observation of a number of lensing events occurring in fields in the Galactic Bulge region would reveal information about the Galactic structure in that region and perhaps yield information about the lens masses. The red clump technique is reviewed and applied to several 3x 3 arcminfields observed at low galactic latitude and longitude in the Galactic Bulge. The derived distances to the fields show consistency with the E2 bar model proposed by Stanek, and indicate that the distance from our sun to the Galactic Centre is 7.97± 0.4 kpc. Furthermore, an attempt is made to apply the derived individual distance to each field to calculate the possible lens mass associated with each event.

The polarisation of radio emission is one of the most powerful probes of magnetic fields in the cosmos. Faraday rotation of polarized radiation provides one of the methods to observe magnetic fields. Measuring the rotation of the polarisation angle of radiation from an extragalactic source over a broad radio bandwidth allows us to infer the properties of the magnetic fields that the radiation passed through on the path to the observer. In the last few decades, the presence of structure in the matter distribution of the universe has been observed. It remains an open question whether there are magnetic fields associated with this large-scale structure. Large-scale universe simulations allow us to investigate the effect of extragalactic magnetic fields on the spatial distribution of Rotation Measure (RM) of radio sources that will be detected in deep radio images with MeerKAT. We constructed lightcones out to z = 1 from large-scale universe simulations as a base for our model and assemble a routine to trace large scale structures, attach magnetic fields to the structure and construct RM observations. The aim is to explore whether deep MeerKAT continuum observations will be able to detect magnetic fields associated with large-scale structure (the so-called magnetic cosmic web).

Information on the origin and evolutionary processes of the Solar System is harbored by primitive bodies called trans-Neptunian objects (TNOs). Their preserved state is due to these bodies having orbits at and beyond that of Neptune, as this specific area is considered to be the least thermally modified in the Solar System and could contain a large population of primordial remnants. These archaic remnants not only provide us with information on our infant Solar System, but also improve our understanding of extrasolar planetary formation processes. Stellar occultations by TNOs enable the determination of sizes and shapes with kilometric accuracy (confining albedos, leading us to compositions and densities), the detection of atmospheres down to pressures of a few nanobars, as well as an investigation of the immediate vicinity of the target body (indicating the presence of rings, satellites, jets, comas). These TNOs roam the icy outskirts of our Solar System which, together with their small sizes, make them faint and their stellar occultations short-lived. With a duration ranging from a couple of seconds for small bodies up to a few hundreds of seconds for the larger TNOs, specific imaging cameras combined with Global Positioning devices and optical telescopes, ensure the fast-cadence capture of these events with microsecond timing accuracy. This high-time resolution observing in combination with the relative speed of the star and the occulting body ensures accurate results with resolutions down to the kilometriclevel, which overthrows the resolvability of ground-based telescopes and rivals that of space probes. Here, we present work done by the South African Astronomical Observatory (SAAO) stellar occultation observing program during the period of July 2016 - July 2017. Specifically, we describe the telescopes, instruments and data analysis pipelines that are used for the SAAO stellar occultation program. The stellar occultation results are obtained from the slightly modified, data reducing SHOC pipeline as originally developed by Dr. Marissa Kotze. The SHOC pipeline laid the groundwork for two additional pipelines to be developed and therefore the MORIS instrument on the 3-m IRTF as well as the FLI autoguider cameras mounted on multiple 1-m LCO telescopes are included to provide many opportunities to observe the predicted stellar occultations. These pipelines include reduction features, correct and accurate derivation of timing information, optimization of the signal-to-noise ratio (SNR) through aperture corrected photometry and most important, provide the user with light curve plots of the point sources. The light curves are normalized and individually analyzed for any signs of a positive stellar occultation detection while testing the effects of reduction on the SNR. This is followed by checking the statistical distribution of the data as well as determining a few values for the line-of-sight optical depth. Instrumental deadtimes for the LCO guider cameras are calculated to effectively determine and use time allocated through proposals. Finally, a single positive stellar occultation by Orcus was observed on 7 March 2017 from two separate sites. Here, chord length calculations as well as timing offsets are calculated from the normalized light curves which led to a possible detection of both Orcus and Orcus’s satellite, Vanth. An in depth discussion is provided to justify this reasoning. This thesis serves to characterize and consolidate the now well-established program of stellar occultation observations at the SAAO.

Includes bibliographical references (leaves 47-54).

Observations are presented for 101 I RAS sources that were expected to be Mira Variables and that lie close to the galactic plane: -7.0 3 < b < + 1° and 282.°5 < l < 285.°5. Over 1090 JHKL photometry modules were obtained for the sources. From the photometry the sample was shown to comprise 85 Miras, 12 of which had published periods. New periods were determined for all the Miras. Of the 85 Miras in the sample, 21 had previous published spectra and the rest were tentatively classified as oxygen or carbon-rich Miras using infrared two-colour diagrams. The oxygen and carbon-rich Miras, respectively, have mean periods and K half-amplitudes of 386 days and 0.36 mag and 442 days and 0.46 mag. Statistical tests showed that the probability that the period distribution functions of the oxygen and carbon-rich stars came from the same population was very small: 0.00055. The period as well as the K- Land K- [12μm] colours were shown to be functions of amplitude. For the carbon-rich Miras the [12μm]- [25μm] colour was shown to be correlated with amplitude. Bolometric magnitudes were determined by integrating under a spline curve fitted to the J H K L12, 25μm fluxes of the Miras. Distances to the stars have been calculated using absolute magnitude P- L relations from oxygen-rich and carbon-rich Miras in the LMC, where the LMC distance modulus of 18.57 was adopted. In addition, a search of Miras in the Galactic Bulge using APM-scanned !-plate data are presented. A total of 116 Mira candidates were identified, 43 of which were classified as strong candidates. The presented work will contribute to the study of Galactic structure.

Bibliography: pages 103-117. / Coherent and quasi-periodic rapid oscillations have been seen in the optical and X-ray light curves of more than 30 cataclysmic variables to date. The observational characteristics of these oscillations are reviewed, and current models discussed. To put the observations in context, a detailed review is given of the canonical model of cataclysmic variables and the variations that give rise to its various sub-classes. Observations of two systems of particular interest are presented, and their contributions to the modelling of rapid oscillations are discussed. High-speed photometry of the very bright nova-like variable CPD-48°1577 (IX Velorum) is shown to reveal low-amplitude oscillations with periods between 24 and 31 seconds. The oscillations are present in more than half of the runs, making this variable an attractive target for simultaneous observations in different wavelength bands. The dwarf nova Z Cha is one of four cataclysmic variables which show rapid oscillations as well as primary eclipses. Photometry of z Cha, archived at the University of Cape Town over a period of ten years, is searched for rapid oscillations. Two runs show oscillations, with periods of 24 and 28 seconds. They exhibit eclipse-related phase changes which show qualitative repetition within a run, but significant long-term variation. This behaviour has not been seen in any other cataclysmic variable, and makes z Cha a potentially powerful discriminating tool for models of oscillation and superoutburst.

Includes bibliographical references / As part of a larger effort to uncover the structures hidden behind the Milky Way, we analyse 5,190 spectroscopic redshifts for galaxies in the Hydra/Antlia and Vela regions (245°≤ l ≤ 295°, /b/ ≤ 10°). These galaxies are based on deep optical galaxy source catalogues in the Zone of Avoidance in the Hydra/Antlia region (Kraan-Korteweg 2000a), the Vela region (von Maltitz 2012), and the near infrared 2MASS Extended Source Catalogue(XSC) (Jarrett et al. 2000a, Skrutskie et al. 2006). The observed redshifts were mainly obtained from the 2dF+ AAOmega spectrograph at the Anglo-Australian Telescope (80% of the redshifts); the remaining data are from other telescopes as well as from the literature. This analysis is suggestive of the existence of a massive supercluster in this region, hereafter called the Vela Supercluster (VSC). The prospective VSC is at a mean redshift of cz ~18, 000km s⁻¹, and extends approximately about 87 x 87Mpc on-sky. We use a nearest neighbour algorithm to identify the galaxy clusters and groups within the potential VSC, determine their velocity dispersions and the corresponding virial masses. Although the VSC is sparsely sampled, we find 13 galaxy clusters and 22 galaxy groups contained within it. The richness and the masses of these galaxy clusters/groups are comparable with the galaxy clusters/groups found within the Shapley Supercluster by Proust et al. (2006). Taking account of the sparse sampling, the VSC seems comparable to the Shapley Supercluster (SSC). Given the fact that the SSC contributes about 9% of the Local Group motion (Muñoz & Loeb 2008), if not more (see Kocevski & Ebeling 2006). The proximity of a further supercluster (VSC) in its vicinity, might have serious implications on the bulk flows studies. Consequently, the existence of this supercluster is likely key in resolving the long-enduring bulk flow controversies and the misalignment of flows with the dipole determined via the Cosmic Microwave Background (CMB) observations.

The numerical investigation of the oscillations of white dwarfs which is discussed below was begun at the suggestion of Professor Warner soon after the discovery of periodic variations in several cataclysmic variables. Its completion within a reasonable time was greatly helped by the availability of a computer program for the solution of the equations governing the structure of white dwarf stars provided by Dr W. B. Hubbard. Although the program could not be used in the original form for the present investigation, where the accuracy of the outer layers is critical, it provided an invaluable starting point and the coding of the equation of state of the core was used in an almost unaltered form.

Includes bibliographical references (p. 97-105).

Includes abstract. / Includes bibliographical references. / This dissertation presents a survey made in the 21 cm wavelength line of the neutral hydrogen (HI) to determine redshifts of 928 galaxy candidates in the Zone of Avoidance (ZoA). The 2MASS Redshift Survey (2MRS) is currently the most uniform “whole-sky" redshift survey for mapping large-scale structures, and studying the dynamics in the nearby Universe and the Cosmic Microwave Background (CMB) dipole. While this survey is supposed to cover the whole sky, it excludes the inner ZoA because of difficulties in detecting galaxies due to the heavy obscuration caused by dust and stellar crowding. The lack of redshift data in the ZoA remains the main source of uncertainties in studies of the nearby Universe. To improve on this, the 100m-class Nançay Radio Telescope was used to measure the 21 cm line emission of 928 near-infrared bright galaxy candidates without redshift information in the ZoA which are accessible from the NRT (Dec > -39°), hence focuses predominantly on the mostly unexplored northern ZoA.