Multi-Decadal Variability of West African Rainfall in CMIP5 Simulations

Rainfall in the Sahel region of Africa shows a large multi-decadal signal, transitioning from wet conditions in the 1950s and 60s to dry conditions in the 70s and 80s and appearing to recover back to wet conditions during the 90s and 00s. This multi-decadal signal has been attributed to SSTs in the Atlantic (associated with the Atlantic Multi-decadal Oscillation - AMO) and the Indian Ocean. This study uses output from the 20th century (historical) CMIP5 modeling simulations to investigate the ability of the CMIP5 models to capture multi-decadal variability of rainfall in the Sahel. Initial results indicate that the majority of models produce multi-decadal variability in the North Atlantic SSTs (AMO) as seen in observations. While these models produce a multi-decadal signal in Sahel rainfall, the amplitude of this signal is weak, with errors in the spatial structure evident when compared to observations. Further investigation into model fidelity and mechanisms affecting multi-decadal variability in the historical CMIP5 simulations will be shown in order to aid in the interpretation of the CMIP5 decadal prediction simulations.