The spontaneous emission of gravity waves (GWs) from jet stream imbalances is an important source of these waves in the atmosphere. However, its generation and propagation process is not completely understood so far and no fully satisfactory parameterization schemes exist. We use a finite-volume model to study the spontaneous emission of GW from baroclinic waves in the differentially heated rotating annulus experiment. Possible regions of GW activity are identified by the horizontal velocity divergence. Results indicate an increased GW activity within the baroclinic wave and close to the inner cylinder wall. While the former seems to originate in part from spontaneous GW emission from the baroclinic wave, the latter is attributable to boundary layer instabilities. In order to gain a further understanding of the GW source mechanism a tangent-linear model for the ageostrophic flow component has been developed. The Boussinesq equations have been linearized about the time dependent geostrophic background flow which is obtained from potential-vorticity inversion. They are forced by the nonlinear terms arising exclusively from that geostrophic flow. To avoid boundary layer instabilities this analysis has also been applied to Boussinesq flow in a horizontally doubly periodic cartesian domain, relaxed to a baroclinically unstable hyperbolic tangent temperature profile. First results for this setup show a high correlation between the horizontal divergence of the full and the linear model. Furthermore, there seems to be a significant internal forcing of the GWs by the balanced flow. Ongoing work that we will also report on uses an empirical stochastic parameterization for the nonlinear self-interactions of the ageostrophic flow. The application of the linear analysis on the GW features in the differentially heated rotating annulus will be discussed as well.

---

*email: hien@iau.uni-frankfurt.de
*Preference: Oral