Excitation of the Two Day Wave in the MLT by Waves Emanating from the Troposphere

Abstract:

Mechanistic model experiments will be presented, with the goal of understanding the excitation mechanism and interannual variability of the quasi two day wave (Q2DW) with zonal wavenumber 3. The model is initialized with the observed zonal mean structure of the atmosphere for austral summer solstice for various years. The summer jet contains regions that are baroclinically unstable, in which random wave excitation could stimulate unstable growth of the Q2DW, with rate and magnitude that depends on the variable mean state structure. Unstable modes do exist in linear inviscid model experiments, but they become marginally stable when the damping mechanisms of Newtonian cooling, eddy, and molecular diffusion are included in the model.

In nonlinear model simulations with no imposed wave forcing, synoptic waves spontaneously form off of the tropospheric jet structure, and the resulting waves weakly excite and maintain a Q2DW (along with other waves with the same phase speed with zonal wavenumbers 1-4). With the addition of a rich spectrum of waves forced by latent heating (derived from TRMM observations of rainfall rate), a robust Q2DW with amplitude similar to those observed is excited. The unstable regions in the mean flow still play a role in the ease to which QTDWs are excited: The QTDW first appears near the subtropical barotropically unstable region that is associated with the stratopause QTDW. EP flux of the mature QTDW emanates from the baroclinically unstable region in the midlatitude jet.

Further experiments, employing artificial local 2DW sources centered at various latitudes and altitudes, show that the QTDW is readily excited by any transient wave source with only moderate variation in efficiency. Furthermore, the amplitude and frequency of the QTDW does not strongly depend on which year the model is initialized. Thus a detailed understanding of the QTDW life cycle in a given year will depend on both the formation of the mean flow that will support resonant growth as well as the occurrence of a random wave event that will significantly excite it.