Magnetic Resonance Thermometry (MRT) leverages temperature-dependent magnetic resonance (MR) parameters within MRI machines, finding applications in both research and clinical settings. Despite its wide range of applications, MRT has not been extensively simulated in MR simulators, limiting the ability to cheaply and easily determine the effects of various parameters on MRT. This thesis investigates the feasibility of implementing an MRT technique within an MR simulator to generate temperature measurements that align with physical observations. The study begins by selecting an appropriate MRT technique and MR simulator, followed by a brief characterisation focusing on the accuracy of temperature measurements of the chosen method. Subsequently, the simulator’s output is compared with experimental MRI data, specifically the temperature change data of a cooling cube. Proton Resonance Frequency (PRF) MRT was chosen, along with the MR simulator PhoenixMR. Results of the characterisation of the accuracy of the reconstructed temperature maps indicate that the type of sequence affects the accuracy of the temperature measurements from the MR simulator. Specifically, with a gradient echo sequence, the accuracy of static temperature change maps is greater than with a spin echo sequence; however, this is reversed for dynamic temperature change measurements over time. It was also found that using an exponential decay model within the MR simulator to model heat diffusion yields simulated outputs that agree with experimental temperature data within experimental uncertainty. Hence, it was determined that PRF MRT could be successfully simulated in PhoenixMR.