Optimized kinematic models for the solar magnetic cycle

Modelling the solar magnetic cycle requires the prescription of several poorly-constrained parameters. Accurate simulations are desirable because the state of the magnetic field at cycle minimum can be used to make predictions about the following cycle. Small-scale parameter surveys have previously been performed in this area, but usually not with global coverage of the parameter space. In this thesis, a genetic algorithm is used to systematically search for optimal parameters for 1D and 2D surface flux transport models, with a view to applying the same technique to a kinematic 3D dynamo model. The method successfully obtains good matches with observations when applied to surface flux transport models. However, for more complex models a more efficient method might be needed. Such a method is Bayesian emulation and history matching, so we apply this method to the surface flux transport models and successfully recreate our results. The contributions of individual active regions to the Sun’s axial dipole moment are assessed by simulating the evolution of each region separately from the others. It transpires that a small number of active regions can have a significant effect on the end-of-solar-cycle dipole moment and hence the subsequent cycle. However, the cumulative effect of less important regions should not be ignored. Emergence latitude is the primary property of an active region in determining its contribution to the axial dipole moment. Finally, a discrepancy between the surface evolution in the surface flux transport model and dynamo model is investigated using a simple 2D diffusion model. The difference is due to radial diffusion which is not present in the surface-only model. An improved, yet suboptimal, match is obtained when either the diffusivity in the convection zone is increased, or the field lines of active regions are manually disconnected from the underlying toroidal field. Increasing diffusivity is a means of disconnecting regions from the toroidal field whilst conserving flux. However, it does not yet appear possible to maintain the dynamo with such a strong diffusivity, although a more thorough parameter optimization could solve this problem