This proposal is motivated by the societal need for skillful decadal forecasts of the West African Monsoon and its associated impacts on Hurricane Activity. It is also motivated by the need to increase our confidence in coupled Atmosphere-Ocean General Circulation Models (AOGCMs) used for longer time climate prediction – especially given the lack of model agreement in West African rainfall predictions in this region reported in the 4th Assessment Report of the Intergovernment Panel on Climate Change.The two overarching aims of the proposal are: (1) To evaluate the extent to which coupled ocean-atmosphere models are able to predict decadal variability of the West African climate including its impacts on hurricane activity. (2) To investigate the key mechanisms which explain identified decadal predictability of the West African climate as well as the reasons for model-to-model differences in skill. Succesfully addressing these aims is an important step towards establishing an operational decadal prediction system for use in the West African and Tropical Atlantic regions. The work planned will make extensive use of the “short-term” decadal forecasts made for CMIP5, with most effort given to the 10-year and 30-year hindcasts and forecasts. Objective measures of the skill of all available models will be prepared by comparing the forecasts with observations of West African rainfall, sea surface temperatures and reanalysis datasets. The analysis will highlight the extent to which state-of-the-art AOGCMs are able to predict the West African and Atlantic climate on decadal timescales and whether this depends on the state of the ocean. Efforts will be made to explain the sources of predictability in the most skillful models and to assess the extent to which the level of skill relies on realistic representation of monsoon processes in the West African region. Evaluation of the 30-year forecasts will also be used to assess the potential role of anthropogenic forcing on the WAM by exploring trends in rainfall and related trends in SSTs.