Properties of the mesosphere and lower thermosphere

The variability and systematic variations of the properties of the upper mesosphere and lower thermosphere are probably the least well known aspects of the terrestrial atmosphere. Satellite measurements of this region are very limited and rocket and remote sounding techniques do not provide comprehensive coverage. Progress is being made in theoretical studies of this region, primarily with regard to tidal effects, and some progress is being made in analyzing the relatively sparse experimental data that are available. Turbulence dynamics of the region has been studied by analyzing structure measurements at Kwajalein, wind data from Natal and systematic variations of the turbopause altitude determined from measurements of the diffusive separation of argon. One question that is being raised at this time, and it is appropriate at a time near solar maximum, is the extent of solar activity control of the properties of this region of the atmosphere. The occurrence rates and magnitudes of the turbulent diffusivity in the 70 to 90 km altitude region appear to correlate with solar activity with a time lag, as do also the incidence of aurora and the atomic oxygen green line intensity. Solar cycle dependence has been identified in mean zonal wind speeds in the 65 to 110 km altitude region above Saskatoon and in lower thermosphere temperatures measured at Heiss Island and at St. Santin. Millstone Hill data show that the mean meridional wind changes during a solar cycle. Solar cycle variations have also been detected in the stratosphere and troposphere.     

Neutral temperatures from Thomson scatter measurements: Comparisons with the CIRA(1972)

Neutral exospheric and lower thermospheric (100–130 km) temperatures from Thomson scatter measurements at Millstone Hill (42°N) are compared with CIRA temperatures with a view towards identifying deficiencies in the CIRA and recommending revisions. CIRA models the observed diurnal mean temperatures (T₀) to within 10% over a wide range of solar conditions (75? F_10.7 ? 250), but consistently underestimates the diurnal temperatures with maximum deviations approaching 50% of observed amplitudes (180–240 K) at solar maximum (1200 K ? T₀ ? 1400 K). The observed semidiurnal amplitudes, which lie in the range of 20K–80K, are always underestimated and frequently by more than 50%. In the lower thermosphere, tidal oscillations of temperature of order 20K–40K occur which are not modelled by CIRA. In addition, an analysis of exospheric temperatures at Millstone Hill during a magnetic disturbance indicates a response within 1–2 hours from storm onset, whereas CIRA assumes a 6.7 hour delay. Although some of these deficiences are addressed by the more recent MSIS model, there exists a sufficient data base to recommend several additional revisions to the CIRA temperatures at this time.     

Recent advances in upper atmospheric structure

A possible quantitative explanation of the semi-annual variation in thermospheric density has been obtained in terms of a semi-annual variation in the computed globally averaged vertical energy carried by propagating tides from the lower and middle atmosphere into the thermosphere. The effect is primarily due to seasonal changes in the distribution of water vapor and in the solar declination angle and Sun-Earth distance. An MSIS-83 empirical model of the thermosphere, representing a revision of the earlier MSIS models, has been prepared. The database used covers a wider range of solar activity than previous models and an improved magnetic storm representation is included. Atomic oxygen profiles in the 100 to 160 km altitude region of the auroral thermosphere have been recalculated from measured quenching of N₂(A³∑_u⁺) using the latest laboratory rates and the results are in good agreement with the mean CIRA 1972 profile. A new empirical model of thermospheric variations with geomagnetic activity has been developed incorporating variations with local magnetic time, latitude dependent terms which can vary with the magnitude of the geomagnetic disturbance, and an altitude dependent expression for the equatorial wave. A new index ML, derived from the AL index, has been developed that appears to have promise to represent the variations of thermospheric species with geomagnetic activity. Satellite measured values of solar UV flux, ground-based observations of CaK plages, sunspot numbers and 10.7 cm solar radio flux have been analyzed for temporal variations. Some differences have been identified and the significance to empirical and theoretical upper atmosphere models is discussed.     

The turbulent combustion of a propane-air mixture

This paper models the combustion of a turbulent homogeneous mixture of propane and air within a duct having a stationary one-dimensional mean flow. The Bray-Moss model is applied to the closure of the chemical production terms, using a probability density function (pdf) of the temperature which is chosen as the characteristic variable. Under the conditions chosen for the study, chemical kinetic factors are important and the conventional assumption, that heat release is controlled by turbulent mixing, is not valid. The chemical model of Edelman and Fortune for the combustion of hydrocarbons is used and simplifying assumptions are made which reduce the systems of unknowns to that of the temperature alone. This leads to the introduction of two chemical production terms which are defined respectively in a “delay zone”, where the heat release is modest, and a “combustion zone”. The required equations for the Favre-averaged temperature, turbulence kinetic energy and the mean square fluctuation of the temperature are solved numerically. In the delay zone, a comparison is made between a second order Borghi type closure and the pdf closure. Good agreement is found in the case of relatively small turbulence intensity. It is shown that the pdf formulation does not require the two zones to be spatially distinct. Differing chemical source terms can be discriminated instantaneously by the reaction progress variable and contributions to the average production terms appropriately apportioned by its pdf. Predictions are made of the profiles of mean temperature and mean square fluctuation under different initial turbulence levels.     

Properties of the mesosphere and thermosphere and comparison with CIRA 72

Exospheric temperatures of several reference atmosphere are reviewed and a recommendation is made for the exospheric temperature of a proposed mean CIRA. One of the deficiencies of CIRA 72 and other present thermospheric models is the representation of density changes with geomagnetic activity. This deficiency is illustrated with samples of data. The data show the effects of geomagnetic activity, particle precipitation, a solar proton event, and gravity waves. An empirical model developed from the unique AFGL satellite density data bank using multiple linear regression is reviewed. The present model is for low to moderate solar flux and quiet geomagnetic conditions, but it is planned to extend the model to active conditions. Good progress has been made since CIRA 72 was specified in our knowledge and understanding of the properties of the lower thermosphere, although there are still some unresolved problems. The biggest progress has been made in the theory of tidal effects and of particulate energy deposition and of electrojet heating. On the other hand, it is still not possible to define adequately the systematic variations of the lower boundary conditions of thermospheric models. This is due to lack of knowledge of the systematic variations of the structure properties in the 100 to 120 km altitude region and inadequate information on the mesospheric turbulence profile and variations in the turbopause altitude.