Abstract. Six medium-scale gravity waves (GWs) with horizontal wavelengths of λ H =60-160 km were detected on four nights by Taylor et al. (2009) in the OH airglow layer near Brasilia, at 15 • S, 47 • W, during the Spread F Experiment (SpreadFEx) in Brazil in 2005. We reverse and forward ray trace these GWs to the tropopause and into the thermosphere using a ray trace model which includes thermospheric dissipation. We identify the convective plumes, convective clusters, and convective regions which may have generated these GWs. We find that deep convection is the highly likely source of four of these GWs. We pinpoint the specific deep convective plumes which likely excited two of these GWs on the nights of 30 September and 1 October. On these nights, the source location/time uncertainties were small and deep convection was sporadic near the modeled source locations. We locate the regions containing deep convective plumes and clusters which likely excited the other two GWs. The last 2 GWs were probably also excited from deep convection; however, they must have been ducted ∼500-700 km if so. Two of the GWs were likely downwards-propagating initially (after which they reflected upwards from the Earth's surface), while one of the GWs was likely upwards-propagating initially from the convective plume/cluster. We also estimate the amplitudes and vertical scales of these waves at the tropopause, and compare their scales with those from a simple, linear convection model. Finally, we calculate each GW's dissipation altitude, location, and amplitude. We find that the dissiCorrespondence to: S. L. Vadas (vasha@cora.nwra.com) pation altitude depends sensitively on the winds at and above the OH layer. We also find that several of these GWs may have penetrated to high enough altitudes to potentially seed equatorial spread F (ESF) if located somewhat farther from the magnetic equator.

No abstract

[1] The momentum budget of the migrating diurnal tide (DW1) at the vernal equinox is studied using the Whole Atmosphere Community Climate Model, version 4 (WACCM4). Classical tidal theory provides an appropriate first-order prediction of the DW1 structure, while gravity wave (GW) forcing and advection are the two most dominant terms in the momentum equation that account for the discrepancies between classical tidal theory and the calculation based on the full primitive equations. It differs from the conclusion by McLandress (2002a) that the parameterized GW effect is substantially weaker than advection terms based on the Canadian Middle Atmosphere Model (CMAM). In the region where DW1 maintains a large amplitude, GW forcing in the wave breaking region always damps DW1 and advances its phase. The linear advection largely determined by the latitudinal shear of the zonal mean zonal wind makes a dominant contribution to the phase change of DW1 in the zonal wind compared to the GW forcing and nonlinear advection. However, nonlinear advection is more important than GW forcing and linear advection in modulating the amplitude and phase of DW1 in the meridional wind. The DW1 amplitudes in temperature and winds are smaller than the TIMED observations, suggesting that GW forcing is overestimated in the WACCM4 and results in a large damping of DW1.