Future warming in West Africa will have a detrimental effect on the communities living there. To support assessments of climate change impacts, we propose a method for refining regional temperature projections and demonstrate its application to West Africa for the mid-21st century. Our focus is on characterising uncertainty more comprehensively by considering projections of global warming. We calculate a transformation between a frequency distribution of global warming values derived from the Coupled Model Intercomparison Project phase 5 (CMIP5) models and a broader published probability distribution of global warming developed by the Met Office. The latter draws on perturbed parameter ensembles of simpler climate models to account for uncertainties related to the atmosphere, ocean, carbon cycle and aerosol processes that are not well characterised by the CMIP5 ensemble. Noting that West African warming is highly correlated with global warming in the CMIP5 ensemble, and that a significant portion of the uncertainty in projected West African warming arises from the uncertainty in global warming, we then apply the same transformation to CMIP5-derived distributions for warming in different regions of West Africa. The resultant regional warming distributions have longer tails than distributions estimated directly from the CMIP5 ensemble. Our results imply that CMIP5-based assessments of temperature-sensitive applications may underestimate the probability of large (and small) impacts. Our method could be used to refine temperature projections for other regions of the world in which regional temperature changes are highly correlated with global mean temperature changes.