I. Harris

Thermospheric gravity waves: observations and interpretation using the transfer function model (TFM)

Gravity waves are prominent in the polar region of the terrestiral thermosphere, and can be excited by perturbations in Joule heating and Lorents force due to magnetospheric processes. We show observations from the Dynamics Explorer-2 satellite to illustrate the complexity of the phenomenon and review the transfer function model (TFM) which has guided our interpretation. On a statistical basis, the observed atmospheric perturbations decrease from the poles toward the equator and tend to correlate with the magnetic activity index, Ap, although individual measurements indicate that the magnetic index is often a poor measure of gravity wave excitation. The theoretical models devised to describe gravity waves are multifaceted. On one end are fully analytical, linear models which are based on the work of Hines. On the other end are fully numerical, thermospheric general circulation models (TGCMs) which incorporate non-linear processes and wave mean flow interactions. The transfer function model (TFM) discussed in this paper is between these two approaches. It is less restrictive than the analytical approach and relates the global propagation of gravity waves to their excitation. Compared with TGCMs, the TFM is simplified by its linear approximation; but it is not limited in spatial and temporal resolution, and the TFM describes the wave propagation through the lower atmosphere. Moreover, the TFM is semianalytical which helps in delineating the wave components. Using expansions in terms of spherical harmonics and Fourier components, the transfer function is obtained from numerical height integration. This is time consuming computationally but needs to be done only once. Once such a transfer function is computed, the wave response to arbitrary source distributions on the globe can then be constructed in very short order. In this review, we discuss some numerical experiments performed with the TFM, to study the various wave components excited in the auroral regions which propagate through the thermosphere and lower atmosphere, and to elucidate the properties of realistic source geometries. The model is applied to the interpretation of satellite measurements. Gravity waves observed in the thermosphere of Venus are also discussed.

Full non-linear treatment of the global thermospheric wind system—II. Results and comparison with observations

Abstract Results of the method given in Part I are presented with comparison with previous calculations and observations of upper atmospheric winds. Conclusions are that non-linear effects are only significant in the equatorial region, especially at solstice conditions and that non-linear effects do not produce any superrotation.

Full non-linear treatment of the global thermospheric wind system—I. Mathematical method and analysis of forces

The equations of horizontal motion of the neutral atmosphere between 120 and 500 km are integrated with the inclusion of all nonlinear terms of the convective derivative and the viscous forces due to vertical and horizontal velocity gradients. Empirical models of the distribution of neutral and charged particles are assumed to be known. The model of velocities developed is a steady state model. In Part I the mathematical method used in the integration of the Navier-Stokes equations is described and the various forces are analyzed. Results of the method given in Part I are presented with comparison with previous calculations and observations of upper atmospheric winds. Conclusions are that nonlinear effects are only significant in the equatorial region, especially at solstice conditions and that nonlinear effects do not produce any superrotation.

A spectral approach for studying middle and upper atmospheric phenomena

Abstract We report on the application of a newly developed spectral code to the study of the middle/upper atmosphere. The spectral approach offers conceptual and practical convenience for analyzing the generation and interaction of different components of atmospheric activity through the decomposition of the dynamical fields into components with different zonal wave numbers ( m ). As examples and tests, we obtain solutions for the m = 0 (the mean circulation) and 1 (the diurnal tides) components separately (no mutual interactions), as well as the m = 1 component under the influence of the mean circulation. By simulating gravity wave effects with Rayleigh friction and eddy diffusion peaking near 90 km altitude, the mean circulation thus generated can reproduce the observed mesospheric temperature anomaly under solstice conditions. The computed diurnal tides are in good agreement with results obtained earlier by other authors. The large temperature gradient (associated with the m = 0 component) set up by the mesospheric temperature anomaly under solstice conditions creates a condition favorable for the development of baroclinic instabilities in the mesopause layer, especially near the summer pole. In our time dependent calculation, waves with approximately 4-day period are generated in the m = 1 component, superimposing with the 1-day period tides.

Non-linear three dimensional spectral model of the venusian thermosphere with super-rotation—II. Temperature, composition and winds

Abstract We discuss the results from a three dimensional non-linear spectral model of the Venusian thermosphere with CO 2 , O and He. As described in an accompanying paper (Stevens-Rayburn et al ., 1989, Planet. Space Sci . 37 , 701), an expansion in terms of vector spherical and Fourier harmonics is used to represent the latitude and Local Time dependencies, treated as perturbations of a globally uniform atmosphere taken from the empirical model of Hedin et al . (1983, J. geophys. Res . 88 , 73). A rigid shell super-rotation rate, uniform in altitude, is adopted. Standard heating rates with an efficiency of about 20% are taken from the work of Fox (1988, Planet. Space Sci . 36 , 37). Based on the results from earlier analyses of Pioneer Venus neutral composition data (Niemann et al ., 1980, J. geophys. Res . 85 , 7817; von Zahn et al ., 1980, J. geophys. Res . 85 , 7829), height dependent eddy diffusion coefficients around 3 ×10 7 cm 2 s −1 are adopted, assuming a Prandtl number of 1. Following Bougher et al . (1986, Icarus 68 , 284), wave drag in the form of Rayleigh friction is introduced to slow the horizontal winds and to increase the temperature contrast between day and night. With minimal adjusting of the parameterizations for eddy diffusion and Rayleigh friction, the model is successful in reproducing the major observed thermospheric features: the broad daytime maxima in CO 2 and O, with significantly larger values at dusk than at dawn, and the “saw tooth” like density maximum in He around 05:00 LT. To provide a better understanding of the dynamic conditions influencing the temperature and composition, numerical experiments were carried out. (1) The diurnal variations in He are most sensitive to thermospheric super-rotation, and calculations were performed with different rotation periods from 4 to 8 days. The longer the period is, the larger is the day-night increase in the He density and the shorter is the time delay in the density buildup after midnight. We can fit the data best with a super-rotation period of 6 days. Super-rotation also accounts for the dawn-dusk asymmetries in the major species. (2) Given a globally uniform atmosphere as input, larger heating rates yield larger temperature contrast between day and night. The temperature contrast can also be increased by lowering the rate of energy advection from day to night. Thus, an alternative model was constructed, with lower heating rates (artificially reduced by a factor of 2 from the standard values) and enhanced Rayleigh friction to slow the winds, reproducing the observed variations in the temperature and in the major species CO 2 and O. However, in this case, the winds were too small and produced He variations much smaller than observed. (3) With lower heating rate and enhanced Rayleigh friction, a reduced eddy diffusion coefficient for He increased the day-night variations of this species, but produced also a large decrease in its density from the pole to the equator, which is not observed. From these parametric studies we conclude that lower u.v. heating rates with an efficiency significantly less than 20% cannot reproduce simultaneously the observed diurnal variations in the temperature, in the heavier species and in helium. With this heating efficiency, the Venusian thermosphere is strongly non-linear; our model would not converge if it were much larger.

On the structure and dynamics of the thermosphere

Abstract Thermospheric temperature, composition and wind measurements from the Dynamics Explorer satellite (DE-2) are interpreted using a three dimensional, multiconstituent spectral model. The analysis accounts for tides driven by the absorbed solar radiation as well as energy and momentum coupling involving the magnetosphere and lower atmosphere. We discuss phenomena associated with the annual tide, polar circulation, magnetic storms and substorms.

On the diurnal variations in the temperature and composition: a three-dimensional model with superrotation

Abstract Based on a simplified theoretical interpretation of the composition measurements with the ONMS and OIMS experiments on Pioneer Venus, the conclusion was drawn that the rotation rate of the thermosphere should be close (within a factor of two) to that of the lower atmosphere. A more realistic three-dimensional model of the thermosphere dynamics is now being developed, considering non-linear processes, higher order modes and collisional momentum exchange between the major species CO2, CO and O, which describes the diurnal variations in temperature and composition (Niemann et al., JGR , 1980). The computed horizontal winds are about 300 m/sec near the terminators and poles. Results are also presented from a two-dimensional (quasi-axisymmetric) spectral model which describes the four day superrotation in the lower atmosphere of Venus.

The dynamo of the diurnal tide and its effect on the thermospheric circulation

Abstract In the upper thermosphere and at equatorial latitudes, the observations reveal a dominance of the fundamental diurnal tide, and show a close relationship between the measured zonal winds and ion drift velocities. This raises questions about the relative importance of the E - and F -region dynamos and the role of the ions in accelerating the neutral atmosphere. To interpret the measurements, we discuss results from a multiconstitutent model (including O, N 2 and O 2 ), describing in self-consistent form the interactions between the neutral winds, dynamo electric fields, and ion drifts. Hough modes and spherical harmonics are used to describe the fundamental diurnal tides generated in the lower atmosphere and thermosphere, respectively. The largest contribution to the dynamo electric field is associated with the spherical harmonic component P 1 1 generated primarily by the absorbed e.u.v. radiation. The dynamo electric fields associated with the Hough modes S 1 −2 and S 1 1 , generated by the absorbed u.v. and visible radiation in the lower atmosphere, are relatively small. The computed zonal velocities for the neutrals and ions, with the higher order spherical harmonics P 1 3 and P 1 5 included, are synthesized at low latitudes and compared with the observations. The model reproduces reasonably well the salient features : the magnitudes of the neutral winds and ion drift velocities and the ratio between the two (only the computed phase difference is somewhat larger than observed). We analyze the electric potential in terms of different cut-off altitudes below which the (generating) winds are forced to zero. This leads us to conclude that, contrary to common perception, the dynamo electric fields are generated primarily in the F -region not in the E -region. Moreover, the results show that the dynamo interaction causes the neutral winds to increase by about 50%. Since the dynamo electric field is static, this increase occurs primarily in the curl field of the velocity, and the effect on the temperature and density variations is relatively small.

Some new aspects on the superrotation of the thermosphere

Abstract The motion of the thermosphere with a rotational velocity between 10 and 20 per cent in excess of the Earths rotational velocity has been deduced by King-Hele and his co-workers from the change of the inclination of satellite orbits. To date no completely satisfactorily explanation of the observations has been presented. In this paper it is shown that in the thermosphere there exists a small diurnal mean driving force in the eastward direction. This force has not previously been considered in analyses of superrotation. A critical review of the observations and a theoretical analysis that takes account of both equinox and solstice conditions is presented. In the lower height region, where the great majority of observations were made, it is possible to achieve agreement between observations and a dynamical model. Additional observational data are needed in the isothermal region for a more complete analysis.

Polar thermospheric Joule heating, and redistribution of recombination energy in the upper mesosphere

Kellogg, W. W. (1961, J. Met. 18, 373) suggested that transport of atomic oxygen from the summer into the winter hemisphere and subsequent release of energy by three body recombination, O + O + N2→O2 + N2 + E, may contribute significantly to the so-called mesopause temperature anomaly (increase in temperature from summer to winter). Earlier model calculations have shown that Kelloggs mechanism produces about a 10% increase in the temperature from summer to winter at 90 km. This process, however, is partly compensated by differential heating from absorption of UV radiation associated with dissociation of O2. In the auroral region of the thermosphere, there is a steady (component of) energy dissipation by Joule heating (with a peak near 130 km) causing a redistribution and depletion of atomic oxygen due to wind-induced diffusion. With the removal of O. latent chemical energy normally released by three body recombination is also removed, and the result is that the temperature decreases by almost 2% near 90 km. Through dynamic feedback, this process reduces the depletion of atomic oxygen by about 25% and the temperature perturbation in the exosphere from 10% to 7% at polar latitudes. Under the influence of the internal dynamo interaction, the prevailing zonal circulation in the upper thermosphere (small in magnitude) changes direction when the redistribution of recombination energy is considered. The above described effects are very sensitive to the adopted rates of eddy diffusion. They are also strongly time dependent and are significantly reduced for disturbances associated with magnetic storms.