PhD defense by Per Andersen

Abstract
"In this thesis I explore four major topics of supernova cosmology. The underlying theme throughout is to improve supernovae as standard candles and thereby improve our understanding of dark energy, dark matter, and cosmology.

For the last few decades the property that type Ia supernovae can be used as standard candles has played a crucial part in constraining the cosmological parameters of the standard cosmological model. Despite having played this crucial role in modern cosmology the progenitor system for type Ia supernovae has so far not been fully understood. In order to understand all systematic effects related to observing type Ia supernovae we need the ability to reliably identify which progenitor system resulted in the type Ia supernovae that we are observing. One way to constrain these progenitor models is through the predicted delay times, the time between star formation and explosion of the type Ia supernovae. On that background we investigate two different ways of measuring the type Ia supernova rates and delay times and how these two methods can be reconciled in a new model. The findings are that the new model is in good agreement with current observations and preferred over the current consensus model. However to constrain type Ia supernova rates in highly star forming regions additional observations are needed in this regime.

Type Ia supernovae can be used to probe the local cosmological bulk flow. Some studies find that this bulk flow is in excess of what is predicted by the standard cosmological model. We show that by taking into account survey geometry and sampling effects when computing the linear theory prediction of this bulk flow these large observed flows can be explained. The finding is that survey geometry and in particular sampling effects can shift the linear theory prediction towards larger bulk flows. This defuses the apparent tension between the measured bulk flow and theory.

Peculiar velocities of type Ia supernova can systematically effect the observed redshifts of Ia supernovae which in turn can cause derived cosmological parameters to also shift systematically. With focus on the Dark Energy Survey type Ia supernova sample it is determined that the method applied to correct for the effects of peculiar velocities in other current studies is sufficient for the Dark Energy Survey sample. This correction must be associated with an uncertainty that is then propagated with other sources of systematic uncertainties in the total uncertainty budget.

After the success of the type Ia supernova as a standard candle at lower redshifts, many candi- dates for a similarly useful standard candle at higher redshifts have been proposed. The potential utility of these proposed high redshift standard candles is investigated, in particular if they aid in discerning between standard cosmology and a defined subset of modified dark energy models. The finding is that higher redshift standard candles will not contribute to this effort more than lower redshift counterparts as the modified dark energy models under current observational constraints can not deviate significantly from standard cosmology at these higher redshifts. This conversely implies that any deviation from standard cosmology at higher redshifts will pose a problem not just to standard cosmology, but also to the tested subset of modified dark energy models."