Understanding the turbulent nature of the atmospheric flow is still a subject of considerable scientific interest. The atmosphere is characterized by having energy on all scales of motion, from the 3 dimensional small scale boundary layer turbulence to the global scales of stationary quasi 2 dimensional planetary waves. One of the main difficulties in characterizing this flow is the lack of clearly separated spectral regimes, or spectral gaps in the spectrum, for the flow. The geostrophic and quasi-geostrophic flow of the atmosphere at the large scale was shown by Charney [2] to be equivalent to 2 dimensional flow. This classical description of the atmospheric flow-works remarkably well due to the relatively stable stratification and the atmospheres small scale height. The characteristics of the 2 dimensionality of the flow is reflected in the energy spectrum. As an extension of Kolmogorovs, 1941 (K41) [3] theory to the 2 dimensional case, Kraichnan [4] predicted from scaling arguments that the energy spectrum for 2 dimensional flow should scale with wave vector as Ek ~ k~3. This was shown by Wiin-Nielsen [5], in an observational study, to be the case for the atmosphere. The result is remarkable, from the point of view of the energy transfer, in the sense that the main mechanism for generation of atmospheric waves, namely the baroclinic instability mechanism, is 3 dimensional in its very nature. Furthermore, a main forcing mechanism, release of latent heat in the tropics from cumulus convection, is small scale and also of a 3 dimensional nature. That is the main reason, why numerical forecasting is such a hard task in the tropics. The fact that quasi-geostrophic theory is not valid in the tropics is for a completely different reason, namely that the Coriolis force vanishes at the equator. From a forecasting point of view, quasi-geostrophy is obsolete; diabatic processes and divergencies are important for good forecasting.
展开▼
