Asteroseismic inference on the fundamental properties and magnetic activity of solar-like oscillating stars

Dixon, Alexandra (2021). Asteroseismic inference on the fundamental properties and magnetic activity of solar-like oscillating stars. University of Birmingham. Ph.D.

Text - Accepted Version
Available under License All rights reserved.

Download (4MB) | Preview


Asteroseismology, the study of resonant oscillations of stars, is a vital tool for investigating interior stellar processes and accurately characterising stars. My thesis focuses on using asteroseismology of solar-like oscillators and advanced statistical methods to obtain fundamental properties of stars, including their magnetic activity distribution. I begin by setting up the theory and background of asteroseismology and stellar activity. I then follow with three applications all with the common theme of implementing Bayesian statistical inference.

The first study presents our newly developed method to estimate the latitudinal distribution of near-surface stellar magnetic activity. This makes use of the relative shifts of different asteroseismic frequencies which are sensitive to the level of activity in certain regions on the stellar surface. We were able to reproduce the known active latitude distribution on the Sun from helioseismic data and after applying our methods to data on the solar analogue HD173701 as observed by the NASA Kepler Mission we find that its active bands extend across a wider range in latitude than those on the Sun. The second study builds on the concept of activity-induced frequency shifts to investigate the impact this has on predictions of fundamental stellar properties from stellar modelling pipelines. We find that although for most properties this bias should be small, with less than 0.5% in mass, age estimates can have up to a 5% error. We expect this to increase for stars with the highest and lowest inclination angles and with even stronger magnetic field strengths than those simulated in the study. The final chapter presents the use of machine learning to build a prior which can be used when fitting a background model to a power spectrum in order to directly extract bulk stellar properties. Our method is able to distinguish between red clump and red giant branch stars to predict mass, effective temperature, and global asteroseismic parameters for targets observed by Kepler that are consistent with those of the APOKASC-2 catalogue.

Type of Work: Thesis (Doctorates > Ph.D.)
Award Type: Doctorates > Ph.D.
Licence: All rights reserved
College/Faculty: Colleges (2008 onwards) > College of Engineering & Physical Sciences
School or Department: School of Physics and Astronomy
Funders: Science and Technology Facilities Council
Subjects: Q Science > QB Astronomy
Q Science > QC Physics


Request a Correction Request a Correction
View Item View Item


Downloads per month over past year