Malawi currently suffers from high levels of food insecurity, and its largely rain-fed agricultural sector is susceptible to climatic shocks. Despite this high level of vulnerability, little evaluation has been carried out to determine the quality of climate information available to Malawi or how climate change will impact the quantity or quality of their main food crop, maize. This research first analysed the ability of currently available climate models to replicate the past climate of Malawi to gain a better understanding of any uncertainty in the models going forward. This research made use of Python to compare observed climatic variables to the outputs of 21 Regional Climate Models (RCMs) from the Coordinated Regional Climate Downscaling Experiment (CORDEX) initiative, as well as six ERA-interim driven RCMs and the 11 General Circulation Models (GCMs) which form their boundary conditions. Through this evaluation it was determined that currently available RCMs and GCMs perform similarly well in replicating the climate of Malawi, but RCMs allow for better spatial analysis due to their higher resolution. It was also clear from this analysis that the performance of RCMs in Malawi is highly influenced by their boundary conditions. This comparison highlighted that currently available climate models replicate trends in Malawi's temperature well, but the outputs for precipitation are highly divergent. This new understanding of the quality of climatic information highlights the risks of maladaptation when using climate projections for adaptation programmes which are sensitive to precipitation, including for the agricultural sector. Approximately half of all calories eaten in Malawi are from domestically grown rain-fed maize and the quality and quantity of maize produced is highly dependent on climatic conditions. Projecting the existing RCMs into the future, it is possible to see rising temperature trends across the whole country. Keeping the uncertainty of the models in mind, it was also possible to set out three potential precipitation scenarios based on minimum, mean and maximum projected precipitation rates. Using these climatic conditions as inputs to AquaCrop - a public computer-based agriculture model developed by the Food and Agriculture Organisation (FAO), it was possible to determine that the projected yield of maize grown in Malawi's main maize growing region, Central Malawi, could decrease or increase depending on the precipitation scenario applied. Based on the frequency of model results, it is considered likely that that precipitation rates will increase slightly, which would lead to slight increases in maize yields. However, a minimum precipitation scenario - as shown as possible in the model projections - could lead to yields decreasing by up to 93%, highlighting a significant risk to the food supply in Malawi. Sampling of maize-based food grown and sold in Malawi has highlighted that the crops are often contaminated with dangerous levels of aflatoxin B1 (AFB1), a highly carcinogenic natural toxin. AFB1 is a secondary metabolite of the mould species Aspergillus flavus and A. parasiticus, with A. flavus a widespread contaminant in arable agriculture. Climatic conditions are known to be key factors in determining the growth and spread of these moulds, and the likelihood of AFB1 production. Using locally calibrated maize crop data, RCM outputs for Malawi, and AFLA-maize - an empirical model developed by researchers at the Università Cattolica del Sacro Cuore in Piacenza, Italy, it was possible to project likely changes in the concentration and distribution of AFB1 contamination on Malawi's maize crops. This analysis found that climate change is projected to make pre-harvest conditions in Malawi more favourable to AFB1 contamination of maize crops, with the risk of contamination moving northwards in a warming climate. Finally, this research explored the work that is currently being done or is planned in Malawi to tackle the risks posed from food insecurity and aflatoxin contamination in a changing climate. Several recommendations are made to decrease the vulnerability and impact of this risk, while also improving the resilience of the population.