Geomagnetically trapped protons forming Earth's proton radiation belt pose a hazard to orbiting spacecraft. In particular, solar cells are prone to degradation caused by non-ionising collisions with protons in the energy range of several megaelectron volts, which can ultimately shorten the lifespan of a mission. Dynamic enhancements in trapped proton flux following solar energetic particle events have been observed to last several months, and there is a strong need for physics‐based modelling in order to predict the impact these changes may have on orbiting spacecraft. This thesis addresses the need for physics-based modelling by presenting an investigation into inner proton belt variability with a 3D numerical model created from scratch, and by quantifying the impact that variability has on the solar cell degradation of orbiting satellites.

After a review of background concepts in Chapter 1, Chapter 2 presents a case study on satellites undergoing electric orbit raising to geostationary orbit. The increase in solar cell degradation that can occur during a period of proton belt enhancement is calculated for three example orbits. It is found that a large enhancement can cause an additional degradation in solar cell output power by up to ~5% over the course of orbit raising, and further changes of a few percent are shown to occur based on the choice of trajectory, or for a 50μm change in solar cell coverglass shielding thickness.

In Chapter 3 a physics-based numerical model is constructed, solving for proton phase space density in terms of the first, second and third adiabatic invariants μ, K and L. This chapter demonstrates how key processes can be quantified, including transport via radial diffusion, the cosmic ray albedo neutron decay source and coulomb collisional loss. In Chapter 4, a 2D version of the model is applied to derive proton radial diffusion coefficients for a period of solar maximum. This is achieved by varying parameters controlling the rate of radial diffusion in order to optimise the fit between model and data from the CRRES satellite, under the assumption of steady state. Results are compared with diffusion coefficients derived in other literature, and the validity of the steady state assumption underlying this technique is discussed.

In Chapter 5, the 3D numerical model is applied to investigate time variability at energies of 1-10 megaelectron volts, a crucial energy range for solar cell degradation. Three sets of diffusion coefficients from previous literature are applied to model the time evolution of proton phase space density over the four year period 2014-2018. The sensitivity of modelling results to the choice of diffusion coefficients is discussed, including the effect on the anisotropy of proton pitch angle distributions. In the final research chapter of this thesis, Chapter 6, these modelling results are then applied to calculate solar cell degradation over the modelling period for an example satellite in 1200km inclined circular orbit. This demonstrates the final step in an end-to-end physics-based calculation of solar cell degradation.