Alternative Theories of Atmospheric Teleconnections and LowFrequency Fluctuations

Observational studies have revealed a rich low-frequency structure in the atmosphere. A review of the theories, observations, and model studies of this low-frequency atmospheric variability is presented. On time scales of weeks or longer the atmosphere appears to possess distinct oscillatory behavior in welldefined and persistent "centers of action." This behavior is also an endemic feature of surrogate atmospheric data sets emerging from experiments with complicated climate models. Many theories have attempted to determine the dominant physical processes responsible for the low-frequency variance but have usually failed when compared carefully with observations. For example, simple linear steady state and Rossby wave dispersion theories have been used in an attempt to explain the observed global response to low-latitude perturbation. However, the observed structures of mature anomalies are often quite distinct from the vertical structures of disturbances predicted in these theories. Also, in general circulation and model studies, the sign of the nonlinear response is not simply related to the sign of the forcing as predicted by linear steady state theories. It is argued that the theories fail because either the full three-dimensional complexity of the basic state is not considered or its inherent instability structure is not recognized or is, in fact, ignored. It is shown that three-dimensional instability theory provides a natural generalization and marriage of the zonally averaged instability theory of Charney and Eady and the Rossby wave dispersion theory of Rossby and Yeh. As such, it provides a formalism which may be used to understand a wide variety of atmospheric fluctuations including the locations of eddy flux covariance maxima and storm tracks in both the tropics and extratropics and the generation of blocking, teleconnection patterns, and other quasi-stationary anomaly features. Attention is focused on two particular mechanisms within this formalism for the formation of quasi-stationary low-frequency fluctuations. One of these is the baroclinic-barotropic dipole instability mechanism in which the formation of quasistationary mature anomalies is initiated by the upstream development of mid-latitude eastward propagating dipole wave trains which arise through the combined baroclinic-barotropic instability of the three-dimensional atmospheric flow. The other is the westerly duct mechanism in which the initiation of low-frequency variability is caused by tropical disturbances. According to this hypothesis, the longitudinal variation of the basic state flow near the equator causes a ducting of wave energy generated in the tropics to specific zones in the upper tropospheric westerlies; these zones then act as source regions for the emanation of waves into the extratropics. Furthermore, this duct also acts as a waveguide for extratropical modes propagating into or through the tropics.