| Summary: | <p>Gamma-rays are known to be emitted from some of the most extreme sources in the known Universe. Many of these objects are volatile and variable, and understanding them is the focus of the work contained in this thesis.</p> <p>One such class of objects which are highly variable in gamma-rays are blazars, which are known to be variable on $\sim$ minute timescales at TeV energies. We present a macroscopic emission model to investigate whether magnetic reconnection is a feasible mechanism which can cause this rapid variability, assuming the primary emission mechanism to be synchrotron self-Compton (SSC). We show that in this case reconnection produces synchrotron-dominated flares so in general cannot fit broadband observations, although rapid flares can be produced.</p> <p>Underlying physical processes occurring in astrophysical sources are believed to relate to the functional forms of the probability distribution function (PDF) of their time-series. Gaussian and lognormal PDFs correspond to linear and multiplicative processes, respectively. By simulating artificial light curves of known PDF functional forms, we prescribe a method for calculating the likelihood that the PDF has been correctly measured, applying this to the blazar PKS2155-304.</p> <p>In recent years, observations made using the Fermi Large Area Telescope (LAT) have revealed classical novae (CNe) as gamma-ray sources, yet not every optically discovered CNe is a gamma-ray source. Here, we present simulations of an artificial Milky Way population of CNe with optical properties based of those detected in M31 and gamma-ray properties based on the observed sample. We show that observations are consistent with all CNe being gamma-ray sources and predict that all CNe at distances <8 kpc will be detected in gamma-rays.</p>
|
|---|