The dynamical mechanisms that lead to the year to year persistence of the North Atlantic Oscillation (NAO) are examined using the 40-year
record of the NCEP reanalysis. The NAO appears as the first mode in a
Rotated Principal Component (RPC) analysis of the monthly-mean geopotential height at 200mb. Associated with the NAO, we find diabatic heating anomalies both in the North Atlantic sector and throughout the global tropics. We use a baroclinic diagnostic model to identify the contributions of tropical and North Atlantic heating, as well as the forcing due to stormtrack transients, to the maintenance of the NAO-covariant stationary waves. Our diagnosis
shows that the dominant forcing to these stationary waves comes not from heating or vorticity fluxes but from zonal-eddy coupling. More than half of the stationary wave response is forced by the interaction between the NAO-covariant zonal-mean zonal wind (UBAR) anomalies and the climatological stationary waves. Furthermore, these UBAR anomalies are in turn maintained primarily by momentum fluxes from the NAO-covariant stationary waves.
Theories of the persistence of the NAO generally emphasize either North Atlantic air-sea interactions or some form of remote forcing, and the presence of strong zonal-eddy coupling has important consequences for both hypotheses. Air-sea interaction theories generally assume that zonal wind anomalies in the North Atlantic are forced by local SSTs, primarily through their impact on the
Atlantic stormtrack. However, the presence of strong zonal-eddy coupling obscures the role of local forcing in generating these anomalies, since zonal-eddy coupling is ostensibly a non-local phenomenon. On the other hand, classical Rossby wave theory is usually invoked to explain the influence of remote forcing, and this theory does not apply when zonal-eddy coupling is in effect. The consequences of strong zonal-eddy coupling for NAO persistence
will therefore be examined in detail.