A study of solar quasi-oscillatory signals

We observe quasi-oscillatory behaviour frequently in solar data, for example in the quasi-biennial oscillation visible in magnetic activity proxies and quasi-periodic pulsations in flares. These behaviours have to be fully understood in order for our solar models to be complete. We can assess these behaviours using a host of analysis techniques that are capable of tracking the quasi-periodicities inherent in these data. In this thesis we use Empirical Mode Decomposition on helioseismic data to examine the quasi biennial oscillation and establish that there is a significant risk of spurious detections when used on short duration data. We also use a combination of the Fast Fourier Transform and Wavelet analysis to obtain the prevalence of non-stationarity in quasi-periodic pulsations using GOES data. We find that quasi-periodic behaviours are extremely common in solar data. We determine that the relative amplitude of the quasi-biennial oscillation is correlated with the amplitude of the solar cycle. We obtain evidence of low magnitude non-stationarity in the majority of quasi-periodic pulsations. Finally we conclude that understanding quasi-oscillatory behaviour is essential to form a complete understanding of the Sun, its dynamo, its magnetic field, and all associated behaviours