This study focuses on the complex and chaotic wave attenuation process through which gravity waves deposit their momentum and affect larger scale flows. The region of interest is New Zealand, motivated by the recent DEEP propagating gravity WAVE experiment over New Zealand (DEEPWAVE-NZ) field campaign occurring June-July 2014. Gravity wave attenuation is studied within a nine-week realistic 6-km resolution WRF simulation and five 2-km resolution event simulations, which were extensively compared against aircraft, radiosonde, and satellite observations. The long WRF simulation revealed a lower-stratospheric mountain wave “valve layer” near 17 km, where the vertical penetration of gravity wave s is very sensitive to ambient wind speed.

Mountain wave attenuation mechanisms within this valve layer are studied using relevant diagnostics (e.g. Richardson number, sub-grid scale momentum flux). Wave dissipation occurs along two turbulent zones tilted along the mountain wave phase lines and is also horizontally inhomogeneous. PV conservation is invalidated in these attenuation regions, generating numerous PV banners downstream. The physics of attenuation is investigated within the valve layer and wave spectra/cospectra are examined above and below to see which waves are able to penetrate the layer and if secondary waves are generated.

This valve layer is also apparent in the parameterized gravity wave drag (GWD) fields in the MERRA reanalysis dataset. GWD is quantitatively compared between WRF and the parameterization at all available altitudes. Qualitatively, the best agreement is seen within the valve layer, although the GWD within WRF is 4-5 times larger than that parameterized in MERRA. Such an under-representation of GWD within global climate models would have important implications for simulations of stratospheric and tropospheric climate. We tentatively conclude that the difference arises from the non-linear mountain wave generation process.

---

*email: christopher.kruse@yale.edu
*Preference: Oral