Interannual to Decadal Variability in the Tropical Atlantic

Abstract

Analyses of annual mean sea surface temperatures (SST) from observations for the period 1903‐94 and four different general circulation models (GCMs) were conducted. The two dominant EOFs of all datasets are characterized by two patterns, which are centered in the trade wind zones, at roughly 158N and 158S, respectively. The two patterns are uncorrelated at any lag and the time spectra of the corresponding principle components are consistent with red noise. The SST variability is strongly correlated with wind stress anomalies in the trade wind zones. The correlations between the wind stress and the SST, as well as the correlation between the net heat flux and the SST anomalies are consistent with the assumption that the variability of the upper tropical Atlantic Ocean is forced by the atmosphere. Dynamic feedbacks of the tropical Atlantic Ocean are less important. The variability in the trade wind zones shows a weak correlation with the ENSO mode in the tropical Pacific. Although the database of SST in the tropical Atlantic is as good or better than in the tropical Pacific, the variability of the SST in the tropical Atlantic is not as well understood as in the tropical Pacific. This might be due to the fact that in the tropical Atlantic the variability of the SST is weaker than in the Pacific. The latter is dominated by the El Nino‐Southern Oscillation (ENSO) phenomenon. Although ENSO originates in the tropical Pacific, it affects the global climate. The physical mechanisms responsible for ENSO are well understood and ENSO forecast models have predictive skill up to approximately one year in advance. The same physical mechanism that produces the ENSO mode in the Pacific can also produce an ENSOlike mode in the Atlantic, but due to the different basin geometry it is expected to be much weaker than in the Pacific (Zebiak 1993; Huang et al. 1995; Latif et al. 1996). Analyses of rainfall data over northeast Brazil (Moura and Shukla 1981), a region that frequently experiences drought conditions, suggested that an interhemispheric dipole in the tropical Atlantic SST anomalies has a major impact on the rainfall in this region. The rainfall is strongly related to the position of the intertropical convergence zone (ITCZ) in the early boreal spring, which in turn may be related to anomalous SST patterns.