In this thesis I develop Bayesian approach to model galaxy surface brightness and apply it to a bulge-disc decomposition analysis of galaxies in near-infrared band, from Two Micron All Sky Survey (2MASS). The thesis has three main parts. First part is a technical development of Bayesian galaxy image decomposition package GALPHAT based on Markov chain Monte Carlo algorithm. I implement a fast and accurate galaxy model image generation algorithm to reduce computation time and make Bayesian approach feasible for real science analysis using large ensemble of galaxies. I perform a benchmark test of G ALPHAT and demonstrate significant improvement in parameter estimation with a correct statistical confidence. Second part is a performance test for full Bayesian application to galaxy bulge-disc decomposition analysis including not only the parameter estimation but also the model comparison to classify different galaxy population. The test demonstrates that GALPHAT has enough statistical power to make a reliable model inference using galaxy photometric survey data. Bayesian prior update is also tested for parameter estimation and Bayes factor model comparison and it shows that informative prior significantly improves the model inference in every aspects. Last part is a Bayesian bulge-disc decomposition analysis using 2MASS Ks-band selected samples. I characterise the luminosity distributions in spheroids, bulges and discs separately in the local Universe and study the galaxy morphology correlation, by full utilizing the ensemble parameter posterior of the entire galaxy samples. It shows that to avoid a biased inference, the parameter covariance and model degeneracy has to be carefully characterized by the full probability distribution.

The majority of massive stars end their lives by ejecting their outer envelopes in a corecollapse supernova explosion. The collision of their ejecta with the surrounding circumstellar medium results in the formation of supernova remnants that have been detected at all wavelengths, from radio to gamma-rays. For several dozen supernova remnants, very-long-baseline radio interferometers have spatially resolved the interaction region and directly measured the expansion rates of the shocked gas; many show evidence of the interaction of supernova ejecta with the dense slow winds characteristic of the red supergiant progenitors. Understanding the dynamics and morphology of the interaction region, particularly in young supernova remnants leads to estimates of the total mass of the circumstellar medium, as well as its density distribution around the star given the value of the wind velocity. Here we studied the interaction of the supernova ejecta with different circumstellar environments to investigate the hydrodynamic evolution of young supernova remnants in the SedovTaylor phase. We used the massively parallel, multi-physics magneto-hydrodynamics (MHD) and gravity code, GIZMO, for our simulations. We chose GIZMO for its flexibility in allowing the user to choose different methods to solve the fluid equations, i.e., new Lagrangian Godunovtype schemes, e.g., Meshless Finite Volume (MFV) and Meshless Finite Mass (MFM), as well as various flavors of smoothed particle hydrodynamics (SPH), or Eulerian fixed-grid schemes. Since the majority of previous studies used the latter, we focused on an extensive comparison of all the meshless methods in solving the Sedov-Taylor blastwave test, a problem for which there is an exact solution. For our given compute resources, we found the parameters (e.g., smoothing length, number of neighbours, artificial viscosity, and particle resolution) for each meshless method that gave the best agreement with the exact solution. We then carried out 2D and 3D simulations of the hydrodynamic interaction of the supernova ejecta with varying density profiles assumed for the circumstellar medium, namely: a 1/r 2 density profile, for a typical, spherically symmetric red supergiant stellar wind, and an axisymmetric torus profile, inspired by the observation of a dense, dusty torus of the circumstellar material around the red supergiant, WOH G64 (Ohnaka et al., 2008). Radially assembled Hierarchical Equal Area isoLatitude Pixelization (HEALPix) shells were used to set-up the initial density and velocity profiles for the ejecta, which is marked by a flat inner core and a steeply declining outer edge. The Weighted Voronoi Tessellation code was used to produce the 1/r 2 and axisymmetric torus density distributions. We showed that the growth of Richtmyer-Meshkov instabilities in the 2D and 3D 1/r 2 profiles are visible as early as 20 yrs into the evolution of the remnant and become increasingly unstable up to 100 yr. While 2D simulations of 1/r 2 profiles show the presence of the Richtmyer-Meshkov instabilities in the hot shell of a contact discontinuity, in 3D we see large bubbles and filamentary structure of the instabilities. Our results for the numerical approaches to simulating the systems for the 1/r 2 density cases were broadly consistent with previous studies in the literature where stationary grids were used. Two scenarios with different torus-cavity density contrasts were considered in which we found that the instability rolls along the half-opening angle takes ∼ 40 yr to develop in the axisymmetric torus with smooth density drop, whereas the axisymmetric torus with steep density drop does not develop any instability rolls up to the end of the simulation. We concluded with a discussion of the implications of our models for the morphology of supernova remnants and their expected levels of multi-wavelength emission.

High-resolution spectroscopy of classical T Tauri stars (CTTS) at one micron yields new insight into the interaction of accretion disks, outflowing gas, and the central stars. Eighty-one 0.95–1.12 μm spectra of 38 CTTS were obtained with NIRSPEC on Keck II (R = 25, 000) between 2001 and 2007. They were reduced with a modified version of Redspec that features an improved interface developed by the author to remove intrinsic spectral features from atmospheric calibrator stars. Profiles of the one-micron neutral helium line (λ10830) are powerful probes of the kinematics and geometry of infalling and outflowing gas within the innermost ∼ 10 R∗ of the accreting systems. Subcontinuum blueshifted absorption components, tracing outflow, are found in about 75% of the CTTS and indicate a dual origin for the winds that power outflows observed farther from the star. Modeling of blue absorption and emission indicates that heavily accreting sources, with one-micron veilings rY > 0.5, are dominated by stellar winds, while lightly accreting sources with lower rY show evidence for a mixture of stellar winds and disk winds. Subcontinuum redshifted absorption components, tracing infall, are found in about 50% of the CTTS, almost never when r Y > 0.5, and indicate accretion along magnetic field lines that connect the star to the disk. Modeling of red absorption indicates that in about half of the objects with such features, the absorption morphology is consistent with previously modeled flows, but in the remaining half, consisting of stars with rY ≤ 0.1, wider and more dilute flows are required than have previously been proposed. Over the entire sample, Paγ morphologies are roughly consistent with their expected formation in a funnel flow, but trends with veiling suggest that accretion shocks and winds can also contribute to the profiles. Finally, ratios of r Y to optical veilings are on average higher than the expectations of existing models, indicating the possibility of additional sources of excess emission at 1 μm. The evidence for two types of inner wind and an accretion flow geometry that depends on accretion rate suggests that the means of mass and angular momentum transport in CTTS systems are more diverse than previously realized.

Includes bibliographical references. / This thesis presents a measurement of the Integrated Sachs-Wolfe (ISW) effect through cross-correlation of the Cosmic Microwave Background (CMB) and the galaxy distribution tracing the large scale structure of the Universe. The CMB data used are from the 2013 release from the Planck satellite, and the large-scale structure data are from the 2MASS Photometric Redshift Catalogue (2MPZ). The galaxy data were divided into redshift shells, and HEALPix was used to create pixelised maps of the CMB temperature anisotropies and the galaxy overdensity. The linear galaxy bias, relating the galaxy density distribution to the underlying matter density distribution, was measured through least squares fitting of the theoretical prediction of the galaxy auto-correlation function in each redshift shell.The bias values were then used to rescale the theoretical predictions for the matter-CMB crosscorrelation functions in each shell. The observed cross-correlation function between the Planck and 2MPZ data in each shell was computed, and the uncertainties associated with each measurement were calculated using cross-correlation of simulated CMB and galaxy overdensity maps. To quantify the possible detection of the ISW effect, hypothesis testing was performed through computation of the covariance matrix and χ 2 statistic in each shell. Detection of the ISW effect was found to be preferred over no detection in every case, with a total likelihood ratio of 3.4:1. While this is not quite strong evidence of detection of the ISW effect, this ratio is more than twice better than previous measurements using photometric redshift shells. While the photometric redshifts in the 2MPZ are more accurate than those that have been used before, a strong detection is out of reach with this data, as the redshifts in the 2MPZ are too shallow for more decisive ISW detection.

This thesis presents a first look at the directly measured peculiar velocity low field within the Southern Zone of Avoidance. Large peculiar velocity surveys are hampered by the low detection rates and poor data quality of galaxies in the Zone of Avoidance (ZoA) where the obscuring effects of dust and stars in the Milky Way prevent the detection of galaxies across 10 - 20% of the sky. Moreover, dynamically significant structures lie hidden behind the Galactic plane. Dedicated surveys have been conducted to unveil the mass distribution within the ZoA. The ZoA peculiar velocity survey presented here makes use of deep systematic Hi survey data, new high resolution Hi observations and new deep near infrared (NIR) observations to provide high fidelity measurements for use with the NIR Tully-Fisher relation. Hi observations reveal galaxies where both optical and NIR surveys fail. The HIZOA deep Hi survey conducted at the 64m Parkes telescope revealed ~ 1 000 galaxies in the southern ZoA (Henning et al. 2005, Donley et al. 2005, Shafi 2008) . Accurate Hi linewidths are required for the measurement of Tully-Fisher distances. The fidelity of the Hi linewidths depends both on the velocity resolution and signal-to-noise ratio of the spectrum. New Hi data were therefore acquired at Parkes for 82 galaxies, providing a mean factor ~ 3:4 improvement in the fractional uncertainties in the 50% linewidth. A deep NIR follow-up survey of HIZOA galaxies within 6000 kms -1 was conducted using the 1:4m IRSF telescope using the SIRIUS camera for simultaneous imaging in the near infrared J, H and Ks bands. These deep, high resolution NIR observations are able to penetrate the dust and deblend foreground stars making it possible to detect the NIR counterparts for the Hi-detected galaxies. The survey images have an exposure time of 10 min resulting in a limiting magnitude approximately 2m deeper than the 2MASS survey and the SIRIUS camera has a 00:045 pix -1 pixel scale and 7:07 _ 7:07 field of view. The three-colour images were searched by eye for possible NIR-Hi counterparts resulting in a galaxy catalogue containing 567 galaxies in 422 fields. Of these fields the NIR counterparts were confirmed for 356 Hi galaxies. Algorithms for the subtraction of foreground stars were developed and used to obtain accurate surface photometry of each source. The NIR galaxy colours were used to investigate the nature of extinction in the ZoA. The results show that on average across the southern ZoA, the true extinction is 82% of the DIRBE/IRAS values provided by Schlegel et al. (1998). There is no significant variation in A preliminary peculiar velocity flowfield within the southern ZoA is derived by combining the Hi and NIR data. A strong flow towards the Great Attractor (GA) is observed. This flow field also shows possible indications of backside infall onto the GA, showing that the GA does indeed play an important role in the motion of the Local Group. The success of this study demonstrates the feasibility of further TF peculiar velocity studies in the ZoA, notably the extension of this survey to greater distances and an improved calibration of the TF relation for use in the ZoA.

Bibliography: pages 205-214. / The AM Canum Venaticorum stars are an in homogeneous group of six stars that are characterised by the absence of hydrogen lines in their spectra. The structure of these stars has been the subject of much debate, and single star models were considered as recently as 1992. The prevailing opinion, however, is that the A.Jf eVn stars are semi-detached binaries which consist of a low-mass, degenerate helium secondary which is transferring mass via an accretion disc to a DB white dwarf. Under this model, the orbital periods of the AM CVn systems are in the range 1000 - 3000 s. The AM CVn systems show many of the accretion disc-related phenomena that are observed in non-magnetic (hydrogen-rich) cataclysmic variables. Five of the six systems show low-amplitude optical oscillations with periods between 500 and 2000 s. The oscillations have a complex and variable harmonic structure. Since they are consistent with the predictions of the double-degenerate binary model, the oscillations were assumed to correspond to the orbital period in each system. However, the fundamental periods and their harmonics are unstable, showing amplitude and phase drifts on a time scale of weeks to months. The instability of the photometric periods has therefore been an obstacle to accepting that the AM CVn stars are binary systems. An alternative explanation was proposed by O'Donoghue & Kilkenny (1989): the photometric variations in the AM CVn stars are not their orbital periods, but are caused by the same mechanism that produces superhumps in SU UMa dwarf novae. Superhumps are optical variations observed during the superoutbursts of SU UMa stars with periods a few percent in excess of the orbital period. The superhumps usually disappear as the star declines to quiescence after a superoutburst. Superhumps are thought to be the result of periodic enhancements of tidal dissipation in the outer regions of the accretion disc. In addition to the superhumps in SU UMa dwarf novae, permanent superhumps have been observed in other hydrogen-rich cataclysmic variables. The aim of this dissertation is to present new photometric data and to explore the period structure of four of the six known AM CVn stars to see whether the observations are consistent with a superhump interpretation. The aim is also to draw comparisons between the AM CVn stars and examples of hydrogen-rich cataclysmic variables in which superhumps have been observed, and in this way to strengthen the evidence that similar mechanisms are operating in the hydrogen-rich superhumpers and the multi periodic AM CVn stars.

The study of pulsating stars is a mature and important field of stellar astrophysics. The recent discovery that main sequence stars such as the Sun and the cool Ap stars oscillate with a large number of normal modes has given rise to asteroseismology, a new approach which promises to yield accurate knowledge of the interior structure and dynamics of these stars. Although the techniques of asteroseismology have yet to be perfected, they will provide us with extremely powerful tools to test theories of stellar structure and evolution and to provide detailed knowledge of stellar mass, age, internal rotation, magnetism and convection. They may also provide information on the elemental abundances and mixing and indicate the presence of low-mass companions. In asteroseismological studies, the primary data are the frequencies of the normal modes present in the object of interest. This thesis describes an attempt to perform a definitive frequency analysis of the rapidly oscillating Ap star HD 101065. The results of the intense observing program and the subsequent frequency analysis have been published and we reproduce them here in their entirety. The disadvantage in this approach is that the terseness expected by the editor of a scientific journal is sometimes a stumbling block for the reader not fully acquainted with the field. It is thus the purpose of Part l of this thesis to supplement the papers presented in Part II and the Appendix and to provide a more general background against which they can be read and understood.

We are now firmly in the era of multi-messenger astronomy. The detection of the first binary black hole (BBH) merger in GW150914 [1] opened up the era of gravitational wave astronomy, with a further 9 such mergers being detected during the first two observing runs (O1 and O2) of the LIGO Scientific and Virgo Collaborations (LVC). The first – and currently only – multi-messenger source was detected during O2 and was caused by the merger of two neutron stars in a binary system (BNS) [2]. The electromagnetic (EM) counterparts to GW170817 [3] were observed across the EM spectrum by numerous observing facilities, with implications across a vast range of scientific disciplines. Optical/nearinfrared observations demonstrated that the emission was due to a kilonova powered by the radioactive decay of r-process material produced during the merger. For the first time short gamma-ray bursts were convincingly linked to BNS mergers, as observed in GRB170817A [3]. The third LVC observing run (O3) began 2019 April 1 and concluded 2020 March 27. The signal from GW190425 [4] was likely caused by the coalescence of two neutron stars, with the system having a larger total mass than any currently known BNS system. Furthermore, the detection of GW190412 revealed the first BBH merger with a clearly unequal mass ratio of q = m2/m1 = 0.28 along with significant higher-multipole gravitational radiation [5].

A study was carried out to determine how well the L-Galaxies 2020 semi-analytic model simulates the stellar halos of galaxies and the intracluster stellar (ICS) components of galaxy clusters. Two galaxy disruption models were tested, namely instantaneous disruption and gradual disruption. Furthermore, two stellar halo profiles were applied to the simulation results: a power-law profile with slope γ = −3.5 and a Navarro-Frenk-White (NFW) profile. In the latter case, the stellar halo stars follow the distribution of the galaxy's dark matter. It was found that a combination of an NFW profile and gradual disruption provided the best results across the widest range of literature data, namely measurements of stellar halo mass, total stellar mass, stellar mass fractions, and stellar halo iron abundances. Gradual disruption of satellite galaxies also resulted in the central galaxies having more massive stellar halos in comparison to instantaneous disruption. Additional stellar halo formation mechanisms, such as in-situ star formation, were not needed, as the stellar halo masses seen in observations can be obtained in L-Galaxies by considering only tidal disruption of infalling satellite galaxies. The number of high mass accretions into the halos of Milky Way-mass galaxies in the gradual disruption model agreed well with simulation literature. It was found that while central galaxies can induce many disruptions of satellite galaxies (over a thousand in some cases), the majority of the Milky Way-sized stellar halos in L-Galaxies are formed by the disruption of one to fourteen satellite galaxies, in good agreement with simulation literature. A population of galaxies with unexpectedly low stellar halo iron abundances was found. These were determined to be a result of disruptions of high mass, low metallicity satellite galaxies. Furthermore, rather than iron or oxygen, carbon was found to be the dominant element produced by stellar halo stars for the majority of redshifts in most high mass central and satellite galaxies, due mainly to asymptotic giant branch stars. The relative contribution of stellar halo stars was found to be minor, however, with circumgalactic medium enrichment from halo stars in comparison to outflows from galactic stars being on average . 1%. For clusters with virial masses exceeding 1.6 × 1014M, the brightest central galaxy and ICS (BCG+ICS) stars contained 42.44% of the total cluster stellar iron content, while the fraction MICS MBCG+MICS was found to be 82.50%, both results being in good agreement with observation.

In the ΛCDM model of the Universe, galaxy-galaxy interactions and mergers are considered key drivers in their evolution. These dynamical events lead to peculiar and disturbed morphologies, which can be studied using morphometric statistics. The rotational 2D asymmetry parameter, adopted from optical studies, has recently been used to quantify asymmetries in the H i images of galaxies. This is useful since H i disks typically extend further than the stellar disks and are more sensitive to distortions from interactions. Asymmetries can also manifest in the 1D spectral domain, distorting the shape of the global H i profiles of galaxies. The shape of this profile is determined predominantly by the kinematics of the galaxy, and the H i spatial distribution to a lesser extent. By using archival H i data and simulations, we have begun investigating the systematics and uncertainties of using the 1D and 2D asymmetry parameters for merger studies. In this thesis, we present a new 1D measure of lopsidedness and examine the applicability of two different 2D asymmetry measures. We investigate the evolution of 2D asymmetry of the gas and stellar distributions in a simulated major merger event and demonstrate that the gas distribution registers the interaction before the stellar distribution is affected. We also find that the outer asymmetry of both distributions is considerably higher postmerger, whereas the intensity-weighted asymmetry returns to pre-merger values. We then explore how well the 1D and 2D parameters agree with visual classifications of asymmetry for a sample of 115 WHISP galaxies and observe that the 1D folding difference lopsidedness and the 2D intensity-weighted asymmetry parameters compare well with the visual classification of asymmetries in the H i profiles and images respectively. We examine the relationship between the 1D and 2D asymmetries in WHISP and find that the 1D folding difference lopsidedness and the 2D intensity-weighted asymmetry yield the strongest linear correlation between spectral and morphological asymmetries, with r = 0.53 after inclination cuts have been applied. Lastly, we investigate the location of interacting galaxies in asymmetry parameter space and find that the joint use of 1D and 2D parameters can separate most interacting galaxies from the non-interacting sample.