This study uses models from phase 5 of the Coupled Model Intercomparison Project (CMIP5) to evaluate and investigate Sahel rainfall multidecadal variability and teleconnections with global sea surface temperatures (SSTs). Multidecadal variability is lower than observed in all historical simulations evaluated. Focus is on teleconnections with North Atlantic SST [Atlantic multidecadal variability (AMV)] as it is more successfully simulated than the Indian Ocean teleconnection. To investigate why some models successfully simulated this teleconnection and others did not, despite having similarly large AMV, two groups of models were selected. Models with large AMV were highlighted as good (or poor) by their ability to simulate relatively high (low) Sahel multidecadal variability and have significant (not significant) correlation between multidecadal Sahel rainfall and an AMV index. Poor models fail to capture the teleconnection between the AMV and Sahel rainfall because the spatial distribution of SST multidecadal variability across the North Atlantic is incorrect. A lack of SST signal in the tropical North Atlantic and Mediterranean reduces the interhemispheric SST gradient and, through circulation changes, the rainfall variability in the Sahel. This pattern was also evident in the control simulations, where SST and Sahel rainfall variability were significantly weaker than historical simulations. Errors in SST variability were suggested to result from a combination of weak wind-evaporation-SST feedbacks, poorly simulated cloud amounts and feedbacks in the stratocumulus regions of the eastern Atlantic, dust-SST-rainfall feedbacks, and sulfate aerosol interactions with clouds. By understanding the deficits and successes of CMIP5 historical simulations, future projections and decadal hindcasts can be examined with additional confidence.

A combination of observations and re-analysis was used to investigate the mechanisms of the connection between the Atlantic Multidecadal Oscillation (AMO) and Sahel rainfall. A composite technique based on the AMO index was used to identify differences between warm and cold phases of the AMO. A significant summer rainfall increase over the Sahel during warm phases of the AMO was observed, with large increases during the typical monsoon onset period in June.In spring of warm phases of the AMO prior to monsoon onset, strengthening of the Saharan heat-low and its associated shallow meridional overturning circulation is observed. The intensified shallow meridional overturning circulation increases moisture flux into the Sahel from the south during spring while increased westerly winds from the Atlantic increase westerly moisture flux into the Sahel during spring and summer. The strengthening of the heat-low is accompanied by increases in Mediterranean sea-surface temperatures during warm phases of the AMO that lead to increases in moisture flux convergence in the northeast Sahel. During warm phases of the AMO the African easterly jet is farther north than in cold phases, and increased African easterly wave (AEW) activity across West Africa and into the Atlantic is observed. This increased AEW activity particularly in the early hurricane season, as measured by eddy kinetic energy, may be contributing to the increased number of Atlantic tropical storms during warm phases of the AMO.

A census of 19 coupled and 12 uncoupled model runs from the Intergovernmental Panel on Climate Change (IPCC) Fourth Assessment Report (AR4) shows that all models have the ability to simulate the location and height of the Caribbean low-level jet (CLLJ); however, the observed semiannual cycle of the CLLJ magnitude was a challenge for the models to reproduce. In particular, model means failed to capture the strong July CLLJ peak as a result of the lack of westward and southward expansion of the North Atlantic subtropical high (NASH) between May and July. The NASH was also found to be too strong, particularly during the first 6 months of the year in the coupled model runs, which led to increased meridional sea level pressure gradients across the southern Caribbean and, hence, an overly strong CLLJ. The ability of the models to simulate the correlation between the CLLJ and regional precipitation varied based on season and region. During summer months, the negative correlation between the CLLJ and Caribbean precipitation anomalies was reproduced in the majority of models, with uncoupled models outperforming coupled models. The positive correlation between the CLLJ and the central U.S. precipitation during February was more challenging for the models, with the uncoupled models failing to reproduce a significant relationship. This may be a result of overactive convective parameterizations raining out too much moisture in the Caribbean meaning less is available for transport northward, or due to incorrect moisture fluxes over the Gulf of Mexico. The representation of the CLLJ in general circulation models has important consequences for accurate predictions and projections of future climate in the Caribbean and surrounding regions.

This study assesses projections from 24 CMIP5 models of number, duration, and severity of pluvial and drought events utilizing 6‐month standardized precipitation index. Increased variability of standardized precipitation index is projected globally. More frequent, longer lasting, and stronger pluvials are projected in wet regions, and the same for droughts in dry regions. Worsening pluvials and droughts are most apparent in the Northern Hemisphere midlatitudes and the Americas, respectively. Uniquely, this study investigates pluvials and droughts in locations where the precipitation trend is of the opposite sign. In drying regions, 40% of grid points project an increase in number and 65% project an increase in duration of severe pluvials. Projections for severe drought events in wetting regions show similar projections. As precipitation trends alone do not provide information about pluvial and drought characteristics this study has important implications for planning and resilience.