Properties of convective gravity waves derived by combining global modeling and satellite observations

Gravity waves play the key role in the dynamics of the middle atmosphere. Among different gravity wave sources, convection has been long accepted as one of the most prominent ones. However, due to the broad spectrum of convective gravity waves and limitations of current observation techniques, the contribution of these waves to atmospheric dynamics is still an open issue. Moreover, due to the same reasons, the horizontal and temporal scales of gravity waves forced by convection are not well known. These scales are usually treated in current convective gravity wave parameterizations as free parameters and they are defined by assuming typical scales of convective systems. In this study, we addressed these issues using a unique approach of combining modeling and measurements. In order to determine the scales of convective gravity waves, instead of assuming typical scales of convective systems, a systematic survey varying the spatial and temporal scales as free parameters of the Yonsei convective gravity wave source (CGWS) scheme is performed. Gravity waves are generated using this CGWS scheme and propagated upward using the Gravity wave Regional Or Global RAy Tracer (GROGRAT). Gravity wave momentum flux spectra in terms of horizontal and vertical wave numbers are calculated from simulations and compared with the respective spectrum observed by the High Resolution Dynamics Limb Sounder (HIRDLS). Based on this comparison, combinations of scale sets which reproduce the observed gravity wave spectrum are selected. HIRDLS can only see a limited portion of the gravity wave spectrum due to visibility effects and observation geometry. To allow for a meaningful comparison of simulated gravity waves to observations a comprehensive filter that mimics the instrument limitations is applied to the simulated waves. This comprehensive observational filter takes into account both instrument visibility due to radiative transfer and retrieval as well as the complex observation geometry.[...]