The research presented in this thesis was co-funded by the Engineering and Physical Sciences Research Council (EPSRC) and Roll-Royce plc, as part of the EPSRC Centre for Doctoral Training in Advanced Metallic Systems. This research project was conducted as part of the MUZIC-3 (Mechanistic Understanding of Zirconium Corrosion) consortium, comprised of several universities, laboratories, and industrial sponsors, with the aim of further understanding the corrosion behaviour of zirconium alloys in water-cooled nuclear reactors. This research project aims to further the mechanistic understanding of the corrosion behaviour of zirconium alloys, primarily by exploring the factors affecting oxide grain nucleation and growth, and how they impact overall corrosion performance. While the aim of this project is to focus on the effects of nucleation and growth, other microstructural aspects affecting corrosion performance will also be explored through the application of advanced characterisation techniques. Correlative electron backscatter diffraction (EBSD) has been utilised to compare the crystallographic texture of zirconium oxide with the orientation of the substrate Zr grain from which it formed. This methodology revealed local variations in oxide texture, which were found to correspond to specific metal orientations, due to lattice matching. This behaviour was only observed for specific metal orientations, which were also found to exhibit poorer corrosion performance. These observations were made for two alloy systems: Zircaloy-2 and ZIRLO. The poorer corrosion resistance observed in lattice matching regions was attributed to the formation of an oxide texture that was not conducive for stress minimisation in the oxide layer. Due to the compressive stresses that accumulate in the oxide layer, the growth of large, protective grains is typically limited to those suitably oriented to minimise in-plane stresses. The strict epitaxy inflicted in lattice matching regions prevents the formation of such grains, fundamentally limiting their ability to grow -- thus producing a less protective microstructure. The findings of the aforementioned EBSD study explicitly demonstrate the importance of considering local variations in oxide texture and corrosion performance when utilising high-resolution techniques, such as transmission electron microscopy (TEM). A novel method for the targeted preparation of TEM lamellae using a plasma focussed ion beam (PFIB) system was developed to prevent potentially erroneous comparisons between specimens. This approach utilises complementary EBSD with PFIB specimen lift-out and preparation procedures, to provide a repeatable routine for reliable specimen preparation. Furthermore, this study demonstrated the variation in corrosion rate, oxide texture, and grain size as a result of region-to-region variability through high resolution crystallographic orientation mapping using scanning precession electron diffraction (SPED). To further explore the wider implications and behaviours of region-to-region oxide variability due to lattice matching, a 3D EBSD experiment was performed using a laser PFIB. The volume of material analysed using this method greatly exceeding that previously examined, and provided further insight into the necessary orientations for lattice matching to occur, and how they manifest with local oxide thickness, interface undulations, and substrate grain misorientation. Crucially, it was identified that there is no discreet cut-off point for lattice matching to occur. Rather, a continuous decrease in corrosion rate is observed with deviation from a basal substrate texture -- possibly accommodated by short-scale interface undulations. High-resolution orientation and phase mapping was performed on zirconium oxide films formed at different temperatures, and under the influence of a proton flux, using SPED. A range of specimens were corroded during in-situ proton irradiation experiments at the University of Michigan, using different dose rates to explore the effects of dose-rate and corrosion rate on the resultant oxide microstructure. Quantitative analysis of grain size distribution demonstrated a significant increase in oxide grain nucleation in all of the irradiated specimens although, despite a drastic difference in corrosion rate and dose-rate, no such difference in grain size distribution was observed. This observation indicates that enhanced grain nucleation is not the principle cause for accelerated corrosion in such experiments, and that radiolysis is a likely cause for the enhanced corrosion rates. A similar observation was made for non-irradiated specimens, formed at different temperatures and corrosion rates.