Investigating the origins of thermonuclear supernovae through multi-dimensional radiative transfer modelling

In this work we primarily focus on time-dependent Monte Carlo radiative transfer simulations of Chandrasekhar mass CO WD pure deflagration models and their comparisons with SNe Iax to investigate if the differences can be resolved. We first present 3D synthetic observables, calculated using an approximate NLTE treatment in our radiative transfer simulations (ARTIS-CLASSIC), for an extensive parameter study of single spark pure deflagration models.

These models produce good agreement with bright and intermediate luminosity SNe Iax although some systematic differences persist. Additionally, such models struggle to reproduce the faintest events. We demonstrate that including the contribution from a luminous remnant, predicted by pure deflagration models and suggested to have been observed for SNe Iax, leads to improved agreement with observations.

We also present a simulation of a deflagration model up to 30 days after explosion using an improved NLTE treatment of the plasma in the radiative transfer (ARTIS-NLTE). This represents the first such simulation of a pure deflagration model. Significant NLTE effects are found in the synthetic light curves and spectra which lead to an overall improved agreement with observed SNe Iax.

We finally investigate one alternative explosion scenario for Chandrasekhar mass CO WDs known as the Gravitationally Confined Detonation (GCD) scenario. In this scenario the deflagration is followed by a detonation. We present 3D synthetic observables calculated using ARTIS-CLASSIC for a parameter study of the GCD scenario and compare with observed SNe Ia.

We find that the GCD scenario remains most promising for the over-luminous 91T-like SNe Ia sub-class but does not provide a good match to any of the other SNe Ia sub-classes. Overall the work presented in this thesis strengthens the case of pure deflagration models as the explanation for SNe Iax however differences still remain. Future work should focus on investigating the nature of the luminous remnant in more detail as well as simulating the radiative transfer for both standard pure deflagration models and those with the remnant contribution included utilising a full NLTE treatment of the plasma conditions.