The change in slope of the atmospheric energy spectrum from –3 at synoptic scales to –5/3 at mesoscales was first observed by Nastrom and Gage and it has been explained with many different theories. All theories agree that the –3 slope at synoptic scales is the result of quasi-geostrophic turbulent dynamics. The majority of theories attribute the mesoscale spectral slope to a change in nonlinear turbulent dynamics. An early suggestion, on the other hand, was that linear inertia–gravity waves dominate the atmospheric energy in the mesoscale range – similar to the oceanic case, where linear waves have long been thought to dominate at scales smaller than the geostrophic eddies. In this talk, we present results from a new method that decomposes observed flow into a geostrophic component and a wave component. The method is applied to two data sets: the MOZAIC data set consisting of measurements collected on board of commercial aircraft and data from the START08 campaign, collected by the research aircraft Gulfstream V. In the lower stratosphere, both data sets suggest that the mesoscale flow is consistent with linear inertia–gravity waves. In the upper troposphere, however, the data sets disagree and only the START08 data show consistency with linear waves. This suggests that a reevaluation of the data acquisition and processing is necessary to determine the nature of mesoscale flow in the upper troposphere.

---

*email: joernc@mit.edu
*Preference: Oral