Topological systems are an exciting frontier of condensed matter physics, as well as having deep relationships to quantum computing. Progress continues to be made both towards building practical quantum computers and experimentally identifying topological features in real materials. However, these attempts have to contend with the messiness of the real world including the effects of finite temperatures.

In quantum computing thermal errors are studied through the framework of quantum error correction. This allows a characterisation of how damaging these errors are, and has provided many general results on the limits of how well given quantum codes can endure thermal errors. In the first part of this thesis we study the typical response of one of the best known quantum codes (the two-dimensional toric code) when subjected to thermal noise. This typical behaviour is compared to the limiting bounds and we note differences that arise due to entropic effects.

Topological features are typically properties of the ground state of a system. In some cases this means finite temperatures can be damaging, washing out the topological properties with excitations above the ground state. But in other cases it has been shown that topological systems can have their own unique finite temperature behaviours, which can be used to identify them. In the second part of the thesis we study the topological model Kitaev's honeycomb. Fixing certain conserved quantities restricts the model to special sectors, some of which carry topological phases. We begin by investigating an intrinsically finite temperature property of these topological phases, thermal edge currents. At too high temperatures, close to the bulk gap, the topological behaviour of these currents breaks down. However, we show that at lower temperatures they can be identified and shown to have the expected properties such as robustness to perturbations. Following this we investigate the thermal states of the full model. Our approach to this allows us to probe the topological order of the model in a way that has not been done before and we identify a number of interesting new features of the finite temperature physics.