Large-scale motions of the tropics in observations and theory

Recent observations of the tropical and subtropical atmosphere are interpreted in terms of scaling arguments and wave propagation theory advanced byCharney (1963, 1969).Charney’s idealizations describe the tropical atmosphere in terms of large regions of quasi-nondivergent flow containing small subdomains of heavy convection and divergence, and place emphasis upon the quasi-rotational regions. FGGE (First GARP Global Experiment) observations suggest that strongly divergent local tropical circulations are forced by latent heating and produce important direct modifications of the total wind field. We describe the extent to which the resulting field consists of divergent and rotational components in different analyses of the FGGE data, and present independent supporting documentation of the results in terms of heating estimates and rainfall observations. Local tropical heating rates on the order of 10°C/day are apparently due to latent heat release associated with precipitation rates as large as 6 cm/day during extended periods. The large contribution of the divergent wind is generally underestimated in models that do not retain such energetic local forcings, and this deficiency may be related to general underestimation of tropical-extratropical connections of many linear models. Such connections are commonly cited in relation to El Niño events, the Southern Hemisphere stationary-wave pattern, and in FGGE studies, but are not well simulated in most linear theories. It is not yet clear whether this is an inherent limitation of linear models, or whether the linear models have not yet explored all the potentially relevant ambient states. We explore the latter possibility by construction of a basic state that allows reasonable latitudinal evolution of the wave field. This basic state has zero absolute vorticity gradient throughout the tropics, and deviations linearized about this state are dynamically analogous to a “local” Hadley cell. To the extent that it is appropriate to regard the results in terms of wave propagation, our analysis suggests a prominent role for gravity-inertia waves in the tropics and for the extratropical connections. The relevance of gravity modes to observations and the theoretical explanation of the flat vorticity field remain to be established.