The forced runs of the OGCM generated SSTA variability time series comparable to observations. While the Atlantic-only hybrid coupled model generates damped oscillations of SST and wind stress anomalies, similar to the findings of Zebiak (1993), the Pacific-Atlantic hybrid coupled model generates interannual variability comparable in magnitude with observations. The fact that the correlation is still poor compared to forced runs suggests that other sources of forcing in the Atlantic are also important, or that our coupling processes have inadequacies. The Pacific+Atlantic and the Atlantic-only uncoupled hybrid experiment showed results which were similar to forced experiment with observed wind stresses, but which were more confined to the equatorial region. The Pacific-only uncoupled hybrid experiment generates oscillations of SST and wind stress anomalies, but did not capture the temporal variability of SSTA over the tropical Atlantic. The results of this work are suggestive that Pacific SSTA variability is necessary for the coupled model to generate sustained oscillations over the tropical Atlantic. Yet, Pacific SSTA variability alone does not explain the observed SSTA over the equatorial Atlantic. It is thus speculated that while atmospheric teleconnections between the Pacific and the tropical Atlantic are an intrinsic component of tropical Atlantic variability, with the Pacific ENSO providing a source of disturbances necessary to maintain the otherwise damped interannual variations of SST and wind stress over the tropical Atlantic, local interactions are essential to explain the observed variability of the coupled ocean-atmosphere system over the tropical Atlantic.
Supplementary URL: