Search

Convectively triggered waves are the main driver of the tropical stratospheric circulation. In atmospheric models, the model’s resolution limits the length of the simulated wave spectrum. In this study, the authors compare the tropical tropospheric wave sources, their projection on the wave field in the lower stratosphere, and the circumstances of their upward propagation in the atmospheric model ECHAM6 with three spectral truncations of T63, T127, and T255. The model internally generates the quasi-biennial oscillation (QBO), which dominates the variability in the tropical stratosphere. This analysis focuses on two opposite phases of the QBO to account for the influence of the background wind field on the wave filtering. It is shown that, compared to the high-resolution model versions, the T63 version has less convective variability and less wave momentum in the lower stratosphere at wavenumbers larger than 20, well below the version’s truncation limit. In the low-resolution version, the upward propagation of the waves is further hindered by the highly active (relative to the high-resolution versions) horizontal diffusion scheme. However, even in the T255 version of ECHAM6, the convective variability is too small compared to TRMM observations at periods shorter than 2 days and wavelengths shorter than 1000 km. Hence, to model a realistic tropical wave activity, the convective parameterization of the model has to improve to increase the day-to-day precipitation variability.

In order to simulate the quasi-biennial oscillation (QBO) with a realistic period and amplitude, general circulation models commonly include parameterizations of small scale gravity waves (GW). In this work, we explore how different GW parameterization setups determine the response of QBO properties to a warmer climate. Atmosphere-only experiments in both present day and warmer climate serve as testbed to analyze the effect of four different GW parameterization setups, active in the tropics. Having tuned the GW parameterizations to produce a realistic QBO in present day climate, we analyze changes of QBO properties in the warmer climate. The QBO period decreases in two parameterization setups by similar to 30 %, while the QBO period remains unchanged in the remaining two parameterization setups. In all parameterization setups, the QBO amplitude in the warmer climate weakens below 10 hPa but shows different behaviour above 10 hPa. We show that changes in QBO amplitude and changes in QBO period are inconsistent among experiments. In the chosen experimental design, the inconsistent future change in QBO properties among the suite of experiments depends solely on the choice of the GW parameterization setup.

International audience; We analyze the stratospheric Kelvin and Rossby-gravity wave packets with periods of a few days in nine high-top (i.e., with stratosphere) models of the fifth Coupled Model Intercomparison Project (CMIP5). These models simulate realistic aspects of these waves and represent them better than the tropospheric convectively coupled waves analyzed in previous studies. There is nevertheless a large spread among the models, and those with a quasi-biennial oscillation (QBO) produce larger amplitude waves than the models without a QBO. For the Rossby-gravity waves this is explained by the fact that models without a QBO never have positive zonal mean zonal winds in the lower stratosphere, a situation that is favorable to the propagation of Rossby-gravity waves. For the Kelvin waves, larger amplitudes in the presence of a QBO is counter intuitive because Kelvin waves are expected to have larger amplitude when the zonal mean zonal wind is negative, and this is always satisfied in models without a QBO. We attribute the larger amplitude to the fact that models tuned to have a QBO require finer vertical resolution in the stratosphere. We also find that models with large precipitation variability tend to produce larger amplitude waves. However, the effect is not as pronounced as was found in previous studies. In fact, even models with weak precipitation variability still have quite realistic stratospheric waves, indicating either that (i) other sources can be significant or that (ii) the dynamical filtering mitigates the differences in the sources between models. © 2014. American Geophysical Union. All Rights Reserved.