Search

The National Center for Atmospheric Research thermosphere-ionosphere-mesosphere-electrodynamics general circulation model (TIME-GCM) is used to study the local time variation of the equatorial O (1S) emission rate observed by the wind imaging interferometer (WINDII) instrument on the Upper Atmosphere Research Satellite for equinox conditions during 1992 and 1993 [Shepherd et al., 1995]. In the evening the airglow emission layer is very bright and descends with time. At about midnight the emission rate is drastically reduced, with a deep midnight minimum. Shortly afterward, the emission rate begins to recover at higher altitudes and increases toward dawn. The TIME-GCM simulations show that if the diurnal tide has sufficient amplitude to penetrate the atomic oxygen layer near 97 km, then it significantly alters the atomic oxygen distribution at low latitudes, producing a strong green line variation similar to the WINDII observations. At latitudes greater than 20°N and 20°S latitude the opposite variation occurs, indicating a global oscillation in the atomic oxygen layer. If the diurnal tide is weak, it does not penetrate the layer and there is only a weak semidiurnal variation of the nighttime green line similar to what is generally observed for solstice conditions.

Simulations from the National Center for Atmospheric Research thermosphere-ionosphere general circulation model constitute controlled numerical experiments which may be used to assess current understanding of Earth's upper atmosphere. Comparisons with a long term data base are particularly valuable in this regard. Accordingly model simulations of geomagnetically quiet and active periods are compared with an observational database from Saint Santin. The simulations and observations are for equinox and northern summer and winter during solar cycle minimum. The observations consist of the diurnal variation of the meridional neutral winds near 300 km; harmonic analysis yielded the mean components and the 24-, 12-, 8-, and 6-hour waves. The model/data comparisons for the diurnal variations are good to excellent: differences are generally ≤ 25 m/s with largest differences typically occurring between 1800 and 0600 UT. In the observations, the diurnal component is approximately 60 m/s in amplitude and 12 hours in phase during quiet periods. These values persist during active periods except in summer when the diurnal amplitude falls to 47 m/s. The model predicts a weaker diurnal amplitude in winter than the observations indicate; it also does not predict the observed decrease of the diurnal amplitude in summer with increasing activity. Harmonic analysis of the data indicated that (1) 12-r and 8-hour waves are important in summer and equinox; in winter the variation is largely diurnal; (2) the semidiurnal and terdiurnal waves vary with season during quiet periods; and (3) the largest effect of geomagnetic activity is in the 12- and 8-hour waves. In the model, the semidiurnal and terdiurnal tides do not vary with season. Further, the effects of varying geomagnetic activity are predominantly in the diurnal component; the semidiurnal and terdiurnal tides in the TIGCM are relatively unaffected in marked contrast to the observations. The differences between the modeled and observed winds illustrate the pervasiveness and importance of variability in the atmosphere: in the high-latitude energy and momentum sources, in the solar forcing, and in the waves that originate in the lower atmosphere and penetrate the thermosphere.

[1] During geomagnetic storms, gravity waves propagate from the polar regions toward the equator heating the thermosphere at 140 km and higher. These gravity waves are produced by Joule heating that occurs at latitudes of 60° and higher. The heating leads to an increase in the density of nitric oxide at 140 km in the thermosphere. On some occasions, the increased nitric oxide diffuses downward to the 110 km level causing the nitric oxide density at that level to increase substantially. Two and a half years (11 March 1998–30 September 2000) of Student Nitric Oxide Explorer (SNOE) observations of nitric oxide were examined to look for occurrences of increased nitric oxide produced by Joule heating initiated gravity waves and to determine how often downflow of nitric oxide occurs. The results of this study show that gravity wave heating occurs frequently, about 12–14% of the time at 40° latitude. For about 50% of these events, downflow of nitric oxide from 140 km to 110–120 km occurs the following day. About 2–3% of the time, gravity waves propagate all the way to the 20°N and S latitude band around the equator. On special occasions, downflow of nitric oxide occurs at the equator as the result of Joule heating occurring in the polar regions. This happened on five occasions during the two and a half year period in 1998–2000.

Nighttime thermospheric winds and temperatures have been measured over Fritz Peak, Colorado (39.9°N, 105.5°W), with a high-resolution Fabry-Perot spectrometer for nearly 17 years between 1968 and 1985. We use data for two specific November periods, (1) 1984, near solar cycle minimum, and (2) 1979, near the peak of solar cycle 21, to illustrate the measured solar cycle variation of nighttime neutral gas temperature and winds over the station. In particular, we present data for November 11, 1979, the day following the day when the solar F10.7 radio flux emission reached its greatest daily value of 367 J (1 J = 1.0×10−22 W m−2 Hz−1). The nighttime measurements, all selected for geomagnetic quiet conditions, show a considerable variation of the thermospheric temperature between solar minimum and solar maximum of nearly 500 K but a relatively minor variation in the thermospheric winds. The recently developed National Center for Atmospheric Research thermosphere-ionosphere-electrodynamics general circulation model (TIE-GCM) is used to simulate the geomagnetic quiet time variation of global thermospheric properties for the two periods and also to perform a time-dependent simulation to calculate the thermospheric variation during November 9–12, 1979, when the solar F10.7 flux varied from 314 J, 367 J, 325 J, and 294 J, respectively. The TIE-GCM histories are used to construct the diurnal variation of thermospheric temperatures and winds as a function of altitude over Fritz Peak Observatory. The aeronomy calculated by the TIE-GCM is also used to predict the diurnal variation and height of the 630-nm volume emission rate. The station processor calculates the emission-weighted height-integrated Doppler line profiles and Doppler shift profiles as would be observed from a ground station. These profiles are reduced into winds and temperatures in a manner similar to the experimental measurements and can be compared with the actual observations. The results show that the TIE-GCM calculated Doppler temperature and winds are in reasonable agreement with the observations for the two periods representing solar minimum and extreme maximum conditions, suggesting that the solar flux model and aeronomic processes in the TIE-GCM can simulate the main thermospheric variations throughout the solar cycle. Furthermore, the time-dependent calculated variations of Doppler temperature and winds also remain in good agreement with the observations of November 11, 1979, which is the day following the maximum solar EUV perturbation of the Earth's upper atmosphere during the entire solar cycle 21. These results show the need for time-dependent calculations when geophysical parameters have large changes during the course of the period of simulation. The station processor results indicate that the altitude of the 630-nm emission varies during the solar cycle by keeping at a near-constant pressure surface, whose geometrical height changes with the solar cycle.

Is the Earth's atmosphere becoming more Venus-like because of increased trace gas emissions from human activity? Venus apparently has a runaway “Greenhouse” where massive amounts of carbon dioxide have caused the atmospheric temperature near the surface to reach 700K with an atmospheric vertical temperature profile that decreases monotonically from the surface to the thermosphere (100 km altitude) where temperatures are on the order 100–250K. In fact, the nighttime thermospheric temperature of Venus is so low, 100K, that it has received the name “Venus Cryosphere.” The Venus atmosphere temperature structure is established by trapped infrared (IR) radiation from CO2 near the surface that raises it's temperature to unbearably high levels, but, the IR radiation that escapes to space from the upper atmosphere is so large that the resulting temperature is very cold.

In the middle atmosphere above 60 km, the electron concentration increases with altitude, reaching values of 10 to the 10th per cu m in the daytime ionospheric E region near 100 km. The electrons are more mobile than the ions and diffuse more rapidly through the neutral atmosphere. The electron diffusion polarizes the medium, causing an electric field to develop that acts to retard the electron diffusion and enhance the conduction current of ions. A global zonally averaged numerical model of atmosheric electricity from the ground to 100 km is used to examine the effect of ambipolar diffusion and the earth's geomagnetic field on the currents and fields in the middle atmosphere. The results show that above about 65 km, ambipolar diffusion generates local electric fields and conduction currents that balance electron diffusion currents. The electric fields and conduction currents are a few orders of magnitude larger than the vertical fields and currents calculated from the downward mapping of the ionospheric potential without taking electron diffusion into account. Ambipolar diffusion does not alter the total current flowing in the global circuit. It is a local effect where enhanced conduction currens flow to balance the electron diffusion current.

Nighttime thermospheric winds and temperatures over Fritz Peak Observatory (39.9°N, 105.5°W) have been determined from the Doppler shifts and line profiles of the (OI) 15876-K (6300 A) line emission. The peak winds measured during several geomagnetic storms are shown to be linearly related to the time rate of change of the equatorial ring current parameter (d(DSt)/dt). During these geomagnetic storm periods the time variation of the measured meridional winds are also strongly correlated with the time rate of change of DSt although the observed winds at midlatitudes may lag variations in this parameter by several hours. The energy input derived from d(DSt)/dt is consistent with the high latitude energy input that is necessary in a thermospheric dynamic model to bring the calculated meridional winds into agreement with the observations. These results suggest that d(DSt)/dt is a useful index for determining the magnitude and time history of the high latitude energy input during both quiet and disturbed geomagnetic conditions for use in global thermospheric dynamic models.

A high-resolution Fabry-Perot spectrometer has been used at Fritz Peak Observatory (39.9 degreeN, 105.5 degreeW) to determine the nighttime thermospheric temperature and winds from November 1973 through February 1975. The thermospheric temperature and winds at a height near the F/sub 2/ peak are determined from the Doppler broadening and shift, respectively, of the atomic oxygen line emission at 15867.852 K (6300.308 A). The experimentally derived temperatures and winds during geomagnetically quiet conditions are compared with a three-dimensional semiempirical model of the neutral thermosphere. This model in turn uses global empirical models of neutral temperature, composition, and electron density to determine the pressure forces and ion drag which generate and modify the wind fields. The large-scale details of measured and calculated nighttime meridional wind components are in general agreement, showing maximum equatorward winds during the summer months. Measured and calculated zonal winds agree for the equinoctial and winter months; however, the measured nighttime zonal winds are westward during summer months in constrast to model calculations that indicate a midnight eastward to westward transition. The measurements also indicate that both the temperature and the equatorward component of the meridional wind measured north of the observing station are greater than those measured southmore » of the station. These positive latitudinal temperature and wind gradients are greatest during the summer months. (AIP)« less

The stable auroral red (SAR) arc is excited primarily by the conduction of heat from the magnetosphere into the ionosphere. The energy heats the ambient F region electrons that in turn excite the atomic oxygen emission at 6300 A within the arc. The major SAR arc properties are controlled by only a few ionospheric and atmospheric parameters, i.e., the topside heat flow rate in the electron gas from the magnetosphere to the ionosphere, the electron number density at the F2 peak, and the exospheric temperature of a model atmosphere. Figures are presented that relate major SAR arc properties to measurable parameters and enable an observer to determine the gross SAR arc characteristics from either ground-based measurements of the (O I) 6300-A emission rate and number density of the F2 peak or satellite measurements of the electron temperature and F2 peak number density. A relationship that approximates the background dissociative recombination contribution to the (O I) 6300-A emission rate is also given.

Nighttime thermospheric winds and temperatures over Fritz Peak Observatory (39.9°N, 105.5°W) have been determined from measurements of the Doppler shifts and line profiles of the OI 15867-K (6300 A) line emission. Measurements have been made during various geomagnetic storm periods from 1973 to 1979. The peak equatorward meridional wind and westward zonal wind during these storms are linearly correlated with the square of the auroral electrojet index, , with correlation coefficients of 0.815 and 0.754, respectively. The measured neutral gas temperature increase is related to the time integral of the index with a correlation coefficient of 0.938.

Calculations in modeling the general ionospheric structure observed during solar minimum and solar maximum suggest that the Hinteregger (1970) solar EUV flux values are adequate for modeling the mid-latitude ionospheric structure during solar minimum. However, these flux values should be doubled for modeling solar maximum conditions. The calculations also suggest that the atomic oxygen density at 120 km was approximately a factor of 2 lower during solar cycle minimum than during solar cycle maximum.

A time-dependent two-dimensional numerical model of the minor neutral constituents in the thermosphere is used to examine the effects of winds in altering the local distribution of NO produced in the aurora. The calculations show that thermospheric winds flowing through regions of enhanced local auroral production produce a “plume” of enhanced NO densities downwind from the aurora. The maximum NO densities calculated with and without horizontal winds differ by a factor of 2.6 for the case consideredindicating the importance of horizontal transport in controlling NO densities within the auroral zone.

Nighttime thermospheric winds and temperatures at F region heights have been measured over Fritz Peak Observatory, Colorado (39.9°N, 105.5°W), with a high-resolution Fabry-Perot spectrometer from about November 1972 to the present. The winds and temperatures are obtained from the Doppler shifts and line profiles of the (O I) 15,867 K (630 nm) line emission. The data obtained during geomagnetic quiet conditions (Ap ≤ 20) have been separated into two groups representing solar cycle minimum conditions (1972–1977) and solar cycle maximum conditions (1978–1979). The monthly variation of the nighttime zonal and meridional winds are presented for both groups. The results show that during solar cycle minimum the zonal winds are predominantly eastward during the winter months at speeds of 50–75 m s−1 and westward during the summer months at speeds of 50–100 m s−1. The meridional winds measured to the north and south of Fritz Peak Observatory are equatorward throughout most of the night at nearly the same speed. The equatorward winds in the summer peak near midnight at about 100–150 m s−1. The equatorward winds in the winter also peak near midnight but with smaller speeds, 50–75 m s−1. During solar cycle maximum the zonal winds are again predominantly eastward during the winter months, reaching peak speeds of 60–75 m s−1 in the early evening hours and decreasing afterward. In the summer the winds are eastward during the early evening hours, shifting to westward near midnight and increasing to a maximum speed of 75–100 m s−1 in the early morning hours. The meridional winds are poleward at ∼20 m s−1 in the early evening hours, shifting to equatorward near 2100 LT and then increasing in magnitude with maximum speeds near or after midnight. There is a considerable difference in the speed of the winds measured to the north and south of Fritz Peak Observatory. During the summer the equatorward winds measured to the north are 150–175 m s−1, whereas the equatorward winds measured to the south are 50–75 m s−1. During the winter the speeds are smaller, 50–75 m s−1 and 25–30 m s−1 to the north and south of the station, respectively. Calculations made with the National Center for Atmospheric Research thermospheric general circulation model for different seasons are in reasonable agreement with the observations during both solar minimum and solar maximum. During solar cycle maximum the observed and calculated meridional wind differences to the north and south of Fritz Peak Observatory are attributed to enhanced auroral heating and ion drag associated with magnetospheric convection north of the observatory. The equatorward driven winds slow south of Fritz Peak Observatory owing to the diminishing influence of high-latitude forcings.

A high-resolution Fabry-Perot spectrometer has been used at Fritz Peak Observatory, Colorado (39.9°N, 105.5°W), to determine the nighttime thermospheric temperatures and winds during geomagnetic storm periods from the line profiles and Doppler shifts of the (O I) 15867-K (6300 A) line emission. Data obtained during four geomagnetic storm periods in 1974 are presented: March 20–22, October 15–20, June 29 to July 8, and April 16–27. The data are compared to temperatures determined from the Ogo 6 global empirical model of neutral composition and temperature and to winds calculated from a semiempirical dynamic model of the neutral thermosphere. The results show considerable variations in the dynamic character of the thermosphere which are also related to the intensity of geomagnetic activity. During geomagnetic storms the nighttime equatorward winds are greatly enhanced from their quiet time values with a maximum measured velocity of 640 m s−1 during a Kp = 9 storm. The zonal winds generally develop a westward component relative to the geomagnetic quiet zonal winds. During intense geomagnetic activity the zonal winds are westward at 100–200 m s−1 in the early evening hours, flowing in the direction of magnetospheric convection and opposite of model predictions. The meridional winds measured south of the site are generally smaller than those measured to the north. Sometimes the meridional winds measured north and south of the station flow in opposite directions. The nighttime neutral gas temperatures are observed to increase from their geomagnetic quiet values during the storm and are in general concurrence with the Ogo 6 model predictions.

We used a high-resolution Fabry-Perot spectrometer at Fritz Peak Observatory (39.9°N, 105.5°W) to determine the nighttime variation of thermospheric temperatures and winds from the line profiles and Doppler shifts of the O I 15,867-K (6300 A) line emission. With data obtained during the period April 1975 through March 1976 we examined the variation of the nighttime neutral gas temperature, zonal winds, and meridional winds at F layer heights throughout the year during solar cycle minimum. We compare our observed temperatures to predictions made by the MSIS (mass spectrometer and incoherent scatter) empirical model of neutral temperature and composition in the thermosphere. The calculated nighttime variation of thermospheric temperature is about 100°–150°K lower than measurements for summer and equinox months and about 50°–75°K lower than measurements for winter months. We compare the observed winds to the zonal and meridional winds, which we calculate from a three-dimensional semiempirical dynamic model of the thermosphere by using the pressure gradients specified by the MSIS model. The calculated nighttime winds are in general agreement with observed winds; both reveal a systematic monthly variation during geomagnetic quiet times. The nighttime equatorward meridional winds are greater in summer than in winter. The zonal winds are eastward throughout most of the night during winter; however, as the season progresses toward summer, a transition develops in the early morning hours from eastward to westward winds. This transition occurs earlier in the night as summer approaches. We use model predictions of the diurnal variations of winds and temperatures to calculate diurnal averages of the zonal and meridional wind components over Fritz Peak Observatory. The seasonal variation of these zonal mean values is presented.

Daily variations in the vertical E × B drift velocity at F region heights have been observed by the incoherent scatter radar facility at Jicamarca, Peru. Near sunset, large upward drifts are observed to raise the F layer to high altitudes (∼500 km); this process significantly depletes the ionization in the lower ionosphere. Owing to the decrease in ion drag the neutral gas accelerates. As the layer descends later in the evening, the ion drag increases; as a result, there is a deceleration of the neutral wind. Vertical velocities resulting from the divergence and convergence of the horizontal neutral winds cause adiabatic cooling and heating, respectively. This dynamic interaction between the neutral and ionized constituents may be responsible for the nighttime neutral temperature perturbation observed near midnight at low latitudes.

Thermospheric winds measurement during geomagnetic storms with Fabry-Perot interferometer from Doppler shift of two 6300 A fringe profiles

The molecular-oxygen number density distribution in the 95- to 120-km region of the atmosphere has been determined from solar-occultation measurements made by the Solrad 8 satellite. The solar intensity in the region 1040–1350 A was monitored by a photometer aboard the satellite during occultation of the sun by the earth. The bulk of the energy in the wavelength interval lies in Lyman α, which is primarily absorbed by molecular oxygen. Therefore the solar-occultation measurements are directly inverted to give the number density distribution of molecular oxygen at the occultation tangent point. The data from 93 occultation scans obtained at midlatitudes and high latitudes in both hemispheres from December 1, 1966, to August 31, 1967, are presented. The results show a seasonal variation of molecular oxygen in the lower thermosphere. In general, the molecular-oxygen number density distribution in the summer is approximately 3 times greater than that in the winter.

Solar occulation data gathered by the French solar instrument on board the OSO-8 satellite have been used to determine the ozone number density distribution between about 50 and 70 km. The solar extinction measurements were made near the Mg II h (2803 A) and Mg II k (2796 A) solar emission lines. The derived ozone number density profiles at 22° and 38°N latitude during July 1975 are presented and compared with other observations and model predictions.

The similarity transformation (ST) defines a new class of robust and stable parametric functions with embedded physical shape information to optimize flexibility in fitting or inverting data. The similarity transformation also permits the extraction of information on the shape of a particular class of physical functions, thereby providing the basis for comparing alternative models and for analyzing the information content of data. We employ these properties of similarity transformations to study differences between state-of-the-art physics-based atmospheric models (the thermosphere ionosphere electrodynamic general circulation model, or TIEGCM) and empirical atmospheric models (Mass Spectrometer Incoherent Scatter, or MSIS) and to investigate the universality of these models; we examine the role of noise in determining acceptable resolution for faithful retrieval of physical properties; and we measure the performance of MSIS-based forward models for inversion of ultraviolet remote sensing of the neutral upper atmosphere. The similarity transform method proves to be a valuable new tool for identifying common and discrepant properties of the models. Further, the ST method shows that TIEGCM and MSISE-90 profiles embody similar shape information and that a suitable ST parameterization can be constructed that approximates profiles from either model to within a few percent accuracy.

Molecular oxygen densities determined from a solar Lyman α occultation experiment on Solrad 8 during December 1966 through August 1967 show a large apparent seasonal variation in the earth's atmosphere near 100 km. The seasonal variation appears at all latitudes that are examined by the occultation technique, and it does not appear to be associated with latitudinal gradients. Also, the data do not show substantial differences between sunrise and sunset.

The role of the thundercloud in the global electric circuit has been considered by many researchers. Thus, Holzer and Saxon (1952) have constructed a simple model of a bipolar thunderstorm. The global models considered provide insight into the atmospheric electric circuit but are restricted, both by various analytical mathematical representations and by computer size, to a grid of about five degrees in latitude and longitude. A need exists, therefore, for the development of a numerical regional model capable of resolving small-scale phenomena so that their coupling into the global-scale circuit can be examined. The construction of a two-dimensional quasi-static numerical model of atmospheric electricity is discussed. The model provides a basis for the calculation of the global electric field and current distribution produced by a single thunderstorm generator. In connection with the calculations, the thunderstorm was defined by a quasi-static current source function which generates a dipole charge configuration.

The earth's thermosphere is that region of the atmosphere above the mesopause (80–90 km) where the neutral gas temperature begins to increase from about 200°K to values as high as 600–2,000°K depending upon solar activity. The dynamics of this region of our atmosphere is primarily governed by solar EUV and UV heating, heat and momentum sources associated with high-latitude auroral processes, upward-propagating tides and waves from the lower atmosphere, and ion drag interactions with the ionosphere. The general properties of the thermosphere and the research on thermospheric dynamics prior to 1979 have been reviewed previously in this IUGG series by Dickinson (1975) and Mayr and Harris (1979). Ionospheric processes and large-scale neutral gas-plasma interactions have been reviewed by Evans (1975), and reviews covering thermospheric composition have been given by Carrignan (1975) and Hedin (1979).

A population of protons with energy of some tens of keV, called the ring current, is found near the equatorial region of the magnetosphere at several earth radii. During the main phase of geomagnetic storms the ring current shifts toward lower L values into the region of the plasmapause, which is characterized by steep gradients in the plasma density. This interaction together with an anisotropic pitch angle distribution leads to ring current instability and the growth of ion cyclotron wave turbulence. As wave energy is dissipated in the ambient electron gas by Landau damping, the plasmapause electron temperature is raised to a few electron volts, and a substantial temperature gradient is created with respect to the ionosphere. The energy transferred to the ionosphere by pitch angle scattering in the low collision frequency region and by heat conduction in the collision-dominated regime raises the ionospheric electron temperature to several thousand degrees. Therefore an appreciable number of electrons in the high-energy tail of the Maxwellian distribution, i.e., electrons with energy greater than 2 eV, exist in the F region of the ionosphere at about 400 km, where atomic oxygen is the dominant neutral gas constituent. Two eV is the threshold for excitation of oxygen atoms to the metastable ¹D level, and these O(¹D) atoms emit 6300-A radiation, the signature of stable auroral red (SAR) arcs. Although the energy input rate required to produce electron temperatures sufficient to cause average SAR arcs is less than 0.1 erg cm−2 s−1, the energy radiated in the red line is only about 0.003 erg cm−2 s−1. Thus an SAR arc is an optical manifestation of a slow release of energy from the magnetosphere during a geomagnetic storm. Energetically it is small in comparison with high-latitude auroral processes.

A number of investigations have been conducted regarding the electrical distortion produced by the earth's orography. Hays and Roble (1979) utilized their global model of atmospheric electricity to study the effect of large-scale orographic features on the currents and fields of the global circuit. The present paper is concerned with an extension of the previous work, taking into account an application of model calculations to orographic features with different configurations and an examination of the electric mapping of these features to ionospheric heights. A two-dimensional quasi-static numerical model of atmospheric electricity is employed. The model contains a detailed electrical conductivity profile. The model region extends from the surface to 100 km and includes the equalization layer located above approximately 70 km. The obtained results show that the electric field and current configurations above mountains depend upon the curvature of the mountain slopes, on the width of the mountain, and on the columnar resistance above the mountain (or mountain height).

Nighttime thermospheric winds and temperatures have been measured over Fritz Peak Observatory, Colorado (39.9/sup 0/N, 105.5/sup 0/W), with a high-resolution Fabry-Perot spectrometer. The winds and temperatures are obtained from the Doppler shifts and line profiles of the (OI) 15,867 K (6300 A) line emission. Measurements made during three large geomagnetic storms, when Kp exceeded 8, have shown large-scale thermospheric waves that are probably generated by impulsive heating at high latitudes. These waves are observed to propagate equatorward over the observatory, and on one occasion a poleward traveling disturbance was observed several hours later. The characteristic features of the observed waves during the three storm periods are described, and the global properties of these waves are examined by using a numerical model of the zonally symmetric thermosphere. The model calculations show reasonable agreement with the observations over Fritz Peak Observatory, thus enabling an estimate to be made of the high-latitude energy input that launched the waves.

Nighttime thermospheric winds and temperatures have been measured over Fritz Peak Observatory, Colorado (39.9°N, 105.5°W), with a high-resolution Fabry-Perot spectrometer during the March 2, 1983, geomagnetic storm. The winds and temperatures are obtained from the Doppler shifts and line profiles of the [O I] 15,867 K (630 nm) line emission. The planetary geomagnetic index, Kp, increased abruptly from 4 to 7 during the beginning of the measuring period near 0000 UT on March 2 and remained at nearly that level throughout the night. During the early evening hours near 1900 LT (0200 UT) the zonal winds measured in both the east and west directions from Fritz Peak Observatory were westward at a speed of about 100 m s−1. The westward winds increased in magnitude during the next 2 hours, with the wind speed east of the station reaching 400 m s−1 and the winds to the west of the station reaching 300 m s−1. The westward winds remained at roughly these speeds until about local midnight (0700 UT), when they gradually decreased to about 50 m s−1 in the east, and the winds to the west of the station shifted to eastward at about 50 m s−1. The meridional winds measured in the early evening hours both to the north and south of the station were poleward at speeds ranging between 50 and 10 m s−1. Near local midnight the wind measured north of the station gradually shifted to equatorward, reaching a maximum speed of about 300 m s−1 near 0200 LT (0900 UT) and then decreasing to about 100 m s−1 near dawn. The wind measured south of the station remained poleward throughout the night at speeds ranging between 25 and 100 m s−1. The neutral gas temperature over Fritz Peak Observatory increased throughout the night from about 1300° to 1600°K. Calculations made with the National Center for Atmospheric Research thermospheric general circulation model suggest that the observed winds over Fritz Peak Observatory (L ≅ 3) are under the influence of an expanded magnetospheric convection pattern during the storm. Enhanced magnetospheric convection at mid-latitudes characteristically produces westward winds during the early evening hours and a strong equatorward surge in the postmidnight winds.

A quasi-steady model of global lower atmospheric electricity is presented. The model considers thunderstorms as dipole electric generators that can be randomly distributed in various regions and that are the only source of atmospheric electricity and includes the effects of orography and electrical coupling along geomagnetic field lines in the ionosphere and magnetosphere. The model is used to calculate the global distribution of electric potential and current for model conductivities and assumed spatial distributions of thunderstorms. Results indicate that large positive electric potentials are generated over thunderstorms and penetrate to ionospheric heights and into the conjugate hemisphere along magnetic field lines. The perturbation of the calculated electric potential and current distributions during solar flares and subsequent Forbush decreases is discussed, and future measurements of atmospheric electrical parameters and modifications of the model which would improve the agreement between calculations and measurements are suggested.

The primary source of ionization in the D region is provided by photoionization of nitric oxide by Lyman alpha radiation. The observation of enhanced electron densities in winter at middle and high latitudes therefore has often been attributed to enhanced nitric oxide densities in this region. This phenomenon is often called winter anomaly or winter anomalous absorption. The problem has been studied by combining an ion chemical model with a two-dimensional time-dependent model of the neutral atmosphere. The model extends from the ground to the base of the lower thermosphere and includes nitric oxide transport from the thermosphere as well as auroral production. The calculated nitric oxide distribution exhibits substantial seasonal and latitudinal variations, which strongly influence the ion chemical composition and simulate the winter anomaly very well. Other aspects of coupling between neutral and ion chemistry are also explored.

The paper presents a model of global atmospheric electricity used to examine the effect of upper atmospheric generators on the global electrical circuit. The model represents thunderstorms as dipole current generators randomly distributed in areas of known thunderstorm frequency; the electrical conductivity in the model increases with altitude, and electrical effects are coupled with a passive magnetosphere along geomagnetic field lines. The large horizontal-scale potential differences at ionospheric heights map downward into the lower atmosphere where the perturbations in the ground electric field are superimposed on the diurnal variation. Finally, changes in the upper atmospheric conductivity due to solar flares, polar cap absorptions, and Forbush decreases are shown to alter the downward mapping of the high-latitude potential pattern and the global distribution of fields and currents.

The effect of resistive heating by radio waves on ionospheric temperatures, electron densities, and airglow emissions is examined by using numerical ionospheric structure and heat balance codes. Two cases are studied: (1) a 3-GHz, 10-GW microwave beam from a proposed satellite power station and (2) 1-MW and 3-MW beams of 15-MHz radio waves launched by the Arecibo antenna. By intent, these two cases have similar intensities and geometries of resistive heating. The most dramatic heating effects are predicted to occur in the E region, where a thermal runaway will take place. The E region electron temperature will increase from 200°K to roughly 1000°K, and the E region electron density will increase by a factor of about 3. In the F region, where thermal conductivity plays an important role, temperature increases of 200°–500°K will appear along magnetic field lines passing through the radio wave beams. Enhanced emissions in airglow and molecular infrared lines will also occur. Radio wave heating, when combined with the diagnostic capabilities of the Arecibo incoherent scatter radar, will generate new opportunities to measure the rates of atomic physics processes and neutral atmosphere temperatures and composition at D and E region altitudes.

A two-dimensional photochemical model which treated the troposphere and stratosphere has been extended to include the mesosphere and lower thermosphere. The model is described, and particular emphasis is placed on the transport and chemistry of nitric oxide. Large amounts of NO are produced in the lower thermosphere both by solar radiation during quiet times and through auroral ionization. Using reasonable descriptions of dynamical processes in the two-dimensional model, substantial amounts of NO produced in the thermosphere can reach the stratosphere, particularly at high latitudes during polar winter. This process provides a coupling between the upper and lower atmosphere which may play a significant role in the photochemistry of odd nitrogen and odd oxygen in the stratosphere.