Interferometer (HRDI) zonal mean zonal wind data. We quantify and interpret differences between previous diurnal and semidiurnal predictions, hereafter G SWM-95, and GSWM-98 results. The revised GW stress parameterization accounts for the most profound changes and leads to seasonal variability predictions that are consistent with diurnal amplitudes observed in the upper mesosphere and lower thermosphere. Unresolved differences between HRDI and other wind climatologies significantly affect MLT tidal predictions.

A quasi‐16‐day wave in the mesosphere and lower thermosphere is investigated through analyses of radar data during January/February 1979 and through numerical simulations for various background wind conditions. Previous workers have examined about 19 days of tropospheric and stratospheric data during January 10–28, 1979, and present conflicting evidence as to whether a large westward propagating wavenumber 1 oscillation observed during this period can be identified in terms of the second symmetric Rossby normal mode of zonal wavenumber 1, commonly referred to as the “16‐day wave.” In the present work we have applied spectral analysis techniques to meridional and zonal winds near 95 km altitude obtained from radar measurements over Obninsk, Russia (54°N, 38°E) and Saskatoon, Canada (52°N, 107°W). These data reveal oscillations of the order of ±10 m s−1 with a period near 16 days as well as waves with periods near 5 and 10 days. These periodicities all correspond to expected resonant frequencies of atmospheric disturbances associated with westward propagating free Rossby modes of zonal wavenumber 1. Numerical simulations are performed which demonstrate that the 95‐km measurements of the 16‐day wave are consistent with upward extension of the oscillation determined from the tropospheric and stratospheric data. Noteworthy features of the model in terms of its applicability in the mesosphere/lower thermosphere regime are explicit inclusion of eddy and molecular diffusion of heat and momentum and realistic distributions of mean winds, especially between 80 and 100 km. The latter include a westerly wind regime above the summer easterly mesospheric jet, thus providing a ducting channel enabling interhemispheric penetration of the winter planetary wave disturbance. This serves to explain the appearance of a quasi‐16‐day wave recently reported in the high‐latitude summer mesopause (Williams and Avery, 1992). However, the efficiency of this interhemispheric coupling may be reduced by gravity wave stress. No significant penetration of the 16‐day oscillation above about 100 km is predicted by the model. Reported signatures of a 16‐day periodicity in ionospheric data therefore require modulation of tidal or gravity wave accessibility to the thermosphere, or perhaps in situ excitation.

A set of numerical experiments have been conducted using the National Center for Atmospheric Research Thermosphere-Ionosphere-Mesosphere-Electrodynamics General Circulation Model (NCAR TIME-GCM) to understand the effects of the quasi-two-day wave (QTDW) on the middle atmosphere horizontal wind and temperature fields. A zonal wavenumber three perturbation with a period of 48 hours and a latitudinal structure identical to the (3, 0) Rossby-gravity mode has been included at the lower-boundary of the model. A response in the middle atmosphere horizontal wind fields is observed with a structure qualitatively similar to observations and other model results. There is also some evidence to suggest an increase in the lower-thermosphere QTDW response due to the interaction with gravity waves. Changes are observed in the zonal mean wind and temperature fields that are clearly related to the QTDW, however it is unclear if these changes are the direct result of wave driving due to the QTDW or are from another source. Evidence for nonlinear interactions between the QTDW and the migrating tides is presented. This includes significant (40-50%) decreases in the amplitude of the migrating tides when the QTDW is present and the generation of wave components which can be tracked back to an interaction between the QTDW and the migrating tides. Clear evidence for the existence of a westward propagating zonal wavenumber six nonmigrating diurnal tidal component which results from the nonlinear interaction between the QTDW and the migrating tides is also presented.

Abstract. The capabilities of the global-scale wave model (GSWM) [Hagan et al., 1995] are extended to include migrating thermospheric solar tides. The GSWM thermospheric tidal forcing parameterization is based on neutral gas heating calculated from first principles in the National Center for Atmospheric Research (NCAR) thermosphere/ionosphere electrodynamics general circulation model (TIE-GCM). This is the first time that a physics-based thermospheric forcing scheme has been used in a model like G SWM. Previous two-dimensional steady state linear tidal models used exospheric temperature measurements to calibrate upper atmospheric tidal forcing. New GSWM results illustrate thermospheric tidal responses that are largely consistent with tides in the TIE-GCM. Diurnal temperature amplitudes increase with increasing solar activity, but there is no analogous diurnal wind response. The thermospheric semidiurnal tide is much weaker than the diurnal tide.Semidiurnal temperature perturbations peak in the lower thermosphere where the semidiurnal forcing maximizes. The new in situ results must be combined with the GSWM upward propagating tide in the lower thermosphere, because the upward propagating components dominate the semidiurnal response throughout the region and the diurnal response below •0130 kin. In situ forcing accounts for most of the diurnal response aloft. Our preliminary evaluation of the GSWM thermospheric predictions is inconclusive. More extensive evaluations are necessary to make a firm assessment of whether the model captures the salient features of the seasonal and solar cycle variability of thermospheric tides.

A steady state two-dimensional linearized model that extends from the ground into the thermosphere and captures the salient features of migrating diurnal and semidiurnal tidal components is used to investigate the comparative importance of the principal sources of these waves. The results, which have previously gone unreported in the literature, demonstrate the nonnegligible effects of atmospheric absorption of solar radiation at infrared (in the troposphere) and ultraviolet (in the stratosphere) wavelengths on mesospheric and lower thermospheric semidiurnal and diurnal tidal fields, respectively. In addition, latent heat release associated with cloudiness or rainfall in the troposphere is shown to be another plausible source of semidiurnal variability in the upper atmosphere. The important effects of these sources on the dynamics of the mesosphere and lower thermosphere emphasize the need to include realistic parameterizations of migrating tides at the lower boundaries of middle and upper atmospheric general circulation models. The results of this investigation also suggest that updated parameterizations of tropospheric tidal forcing are needed to further current understanding of tidal variability in the upper atmosphere. IntroductionMigrating solar tides are global-scale atmospheric waves which propagate westward with the apparent motion of the Sun at periods that are harmonics of a solar day. These waves are forced by absorption of solar radiation throughout the Earth's atmosphere. Ground-based and satellite-borne wind and temperature measurements [e.g., Avery et al., 1989; Manson et al., 1989; Vincent et al., 1989; Clark and $alah, 1991; Morton et al., 1993; Fritts and Islet, 1994; Hays et al., 1993; McLandress et al., 1994] reveal that some of the strongest tidal signatures are found in the mesosphere and lower thermosphere (MLT, -,•80-120 km). The uninitiated find these MLT tidal signatures surprising. They expect small or negligible tides in the MLT, because the region spans the coldest place in the atmosphere where comparatively little solar radiation is absorbed [e.g., Hagan, 1995], so in situ thermotidal forcing is either absent or weak. However, even the earliest models of atmospheric tides established that these strong MLT signatures are manifestations of tidal components forced below the region, which grow in amplitude while propagating upward as neutral atmospheric density decreases [e.g., Chapman and Lindzen, 1970; Forbes and Garrett, 1979; Walterscheid et al., 1980; Aso et al., 1981]. Subsequent to the earliest theoretical and numerical investigations of atmospheric tides cited above, increasing complexities were systematically introduced to two-dimensional linearized steady-state numerical models. These improvements quantified the important effects of mean winds and horizontal temperature gradients [Lindzen and Hong, 1974], viscous dissipation [e.g., Walterscheid et al., 1980; Aso et al., 1981], and the turbulent diffusion of momentum [Vial, 1986; Forbes and Hagan, 1988] as well as heat [Forbes, ...

Abstract. We used the National Center for Atmospheric Research thermosphereionosphere-mesosphere-electrodynamics general circulation model (TIME-GCM) to calculate the variability of the migrating diurnal tide in a January through December 1993 simulation. While TIME-GCM captures the salient features of the latitudinal, altitudinal, and seasonal variability of the migrating diurnal tide, upper mesospheric meridional wind amplitudes are somewhat smaller than those that have been observed from the ground and space. The discrepancies may be attributable to unresolved uncertainties in tidal forcing and/or dissipation in the TIME-GCM. However, our diagnostic simulation suggests that the nonlinear interactions between the migrating diurnal tide and stationary planetary wave i produce measurable nonmigrating diurnal tidal components that modulate migrating diurnal tidal amplitudes and account for significant variability in the upper mesosphere and lower thermosphere.

We present results of a series of numerical experiments for January conditions using a linearized spectral model which includes realistic mean winds and dissipation. These experiments were designed to characterize the propagation characteristics of the (3,0) mixed Rossby‐gravity mode through the middle atmosphere and into the lower thermosphere. Our results suggest that the wave magnitude is extremely sensitive to the zonal mean winds assumed in the calculations. In particular, our results suggest that a comparatively weak eastward stratomesospheric jet during northern hemisphere winter can account for a shift in the resonant frequency of the quasi‐two‐day wave response. Further, the inclusion of a realistic lower thermospheric jet in the summer lower thermosphere sets up a reflecting layer and an associated enhancement of the summer mesospheric wave signature. These features are in keeping with meteor and partial reflection drift radar observations.

Abstract. During gravity wave breaking, heating rates are determined by wave advection, turbulent diffusion, and turbulence dissipative heating. A series of numerical experiments show that the total heating rates can be larg (--• +10 Kh -1) and can cause large local temperature changes. The wave advection causes dynamical cooling in most of the wave breaking region, consistent with previous studies. Nonuniform vertical turbulent diffusion causes strong transient heating in the lower part of the wave breaking region and cooling above. The dissipative heating rate is relatively small compared with those due to the dynamical cooling and turbulent diffusion. In these numerical experiments, zonal wind and temperature perturbations of the diurnal tide and the zonal mean zonal wind and temperature compose the background state for the computation. This is used to examine the idea that temperature inversions, often observed in the mesosphere, are related to the gravity wave and tidal wave interactions. The simulation results show that the large temperature changes in this process can form temperature inversion layers that progress downward with a speed similar to that of a diurnal tide phase speed, which clearly suggests the tidal modulation of the gravity wave and mean flow interactions. Such a process is dependent on season and latitude, because the background state stability varies with season and latitude. The development of the temperature inversion is also affected by the gravity wave characteristics. It is also shown that the local mean wind, wind shear, and chemical species can undergo large changes accompanying the temperature inversion.