The role of remote versus local diabatic forcing for the TEJ variability over West Africa on interannual time scales

The upper-tropospheric Tropical Easterly Jet over West Africa (WA-TEJ), an integral part of the West African Monsoon (WAM) circulation, is strongly correlated with Sahel rainfall changes on interannual and decadal time scales. To better understand the covariability between the WA-TEJ and Sahel rainfall, this study investigates whether the variability of the summer mean WA-TEJ intensity is mainly forced by local WAM rainfall changes or whether the WA-TEJ variability is caused by variations in the remote diabatic forcing which in turn might also drive Sahel rainfall variability .

For this, PUMA, a simple AGCM based on dry dynamics, is driven with summer mean 3D diabatic heating (Q) fields derived from reanalyses/observations. The focus is set on simulating the interannual variability (IAV ) of the WA-TEJ because of the limited temporal availability of reliable forcing data (1979-2016). To disentangle the effects of remote versus local Q anomalies, the global Q field is geographically split up into an African part (local forcing) and rest of the globe (remote forcing) such that Q in one part can vary from year to year while Q in the other part is set constant to climatological values.

The PUMA simulations replicate 64% of the observed IAV of the summer mean WA-TEJ intensity when driven with the full Q variability , meaning that both the local and remote Q vary from year to year. Letting only the remote Q vary, the simulations even explain 67% of the observed IAV which implies that anomalies of the remote Q are the dominant driver of WA-TEJ variability . With the help of idealized experiments in which Q anomalies are restricted to certain regions, ENSO-related Q anomalies in the tropical Pacific are identified as the dominant forcing. La Niña strengthens the WA-TEJ whereas El Niño weakens it. Local Q anomalies associated with the observed IAV of WAM rainfall do not explain the IAV of the WA-TEJ. WAM rainfall anomalies alone can only significantly affect the WA-TEJ if they are comparably strong and arranged in a dipole pattern.

In summary, this study suggests that the observed covariability of WA-TEJ intensity and Sahel rainfall on interannual time scales is not explained by the effect of the local WAM rainfall variability on the WA-TEJ, but by the simultaneous, mainly same-sign effect of remote Q anomalies on both WAM rainfall and WA-TEJ intensity.