In September 1983 a series of HF ionospheric modification exper- iments were conducted in Scandinavia using the heating facility near Tromso, Norway. The purpose of these experiments was to examine the mechanisms by which high-power HF radio waves produce geo- magnetic field-aligned striations (FAS) in the auroral E region. The vast majority of the backscatter observations were made with radars operating at 47 and 144 MHz (STARE Finland). Additionally, limited observations were conducted at 140 (STARE Norway) and 21 MHz (SAFARI). These radars are sensitive to irregularities having scale lengths between 1 and 7 m across the geomagnetic field lines. During periods of full power O-mode heating, striations having peak cross sections of 40 - 50 dBsm are observed. Striations are not detected during times of X-mode heating. When the heater output is varied a corresponding change in the cross section is measured. The magnitude of the change is most pronounced for heater level changes in the range 12.5 to 50% of full power. These cross sections are significantly larger than those measured at midlatitudes using the Arecibo heater ((TURN)10('1) m('2)). This is consistent with theoretical studies which indicate that it is easier to excite short-scale FAS at places where the geomagnetic dip angle is large. The growth and decay times of the striations are frequency dependent. The growths are variable and can range up to a few seconds if the ionospheric conditions are only marginal for striation development. Under more ideal conditions, growth times are 10('1) - 10('2) ms at VHF frequencies and 10('2) - 10('3) ms at 21 MHz. Decay times follow closely with predictions based upon diffusion across field lines. The times range from 1 s at 21 MHz.

Earth, Mars, and Venus formed at about the same time and in the same region of the solar nebula. It is therefore reasonable to assume that their initial composition and tectonic activity were similar. I use this as the major assumption in my study of the early atmospheres of these planets. A primary goal was to estimate the partial pressure of CO(,2) in the early atmospheres. This problem was approached in an inverse sense. I used a one dimensional radiative-convective model of the atmosphere to determine what levels of CO(,2) would be needed to maintain the climatic conditions that are thought to have existed on these planets 4.0 billion years ago. The range of possible atmospheric compositions is large for any one planet, but narrows considerably when the composition must be compatible with conditions on all three planets. From a comparative analysis of the planets, I estimate that 4.0 billion years ago the surface partial pressure of CO(,2) on Mars was between 1.3 and 2.1 bars and its partial pressure on Earth may have been as high as 14 bars. The Earth and Martian atmospheres were very stable and would not have gone into a runaway greenhouse state even if the CO(,2) partial pressure had been equivalent to 100 bars on Earth. On the other hand, because of its proximity to the sun, radiative-convective equilibrium could not be reached on early Venus for CO(,2) partial pressures less than about 3.5 bars. For surface pressures up to 11 bars, the Venus atmosphere was very susceptible to the rapid photodissociation of water vapor and the subsequent escape of hydrogen. At partial pressures greater than 15 bars of CO(,2), the increased albedo of Venus due to Rayleigh scattering dominates and the atmosphere becomes stable against a runaway greenhouse state. This stabilizing effect of Rayleigh scattering in massive CO(,2) atmospheres, along with the relative distance of the Earth and Venus from the sun played an important role in the very divergent evolutions of the terrestrial atmospheres.

Sidebands on the high frequency (HF) waves received from two transmitted waves separated by small frequency intervals were first observed by Gordon and Ganguly in January, 1984 at Arecibo, Puerto Rico. Papadopoulos proposed that HF sidebands would arise from the nonlinear interaction of an ELF wave with the two high frequency waves. The nonlinear mechanism for ELF and HF sideband generation is parametric decay of a high frequency radio wave into a low frequency compressional Alfven wave and a high frequency sideband (Papadopoulos et al., 1982). In 1984 Fejer proposed another nonlinear mechanism (phase modulation theory) for HF sideband generation. Fejer points out that the effect of the ponderomotive force is a reduction in the electron density and an increase in the phase delay which is greater for more powerful waves. Numerical solutions show that the phase delay would relate to the power density of the HF wave. This modulation index theory using the W.K.B. approximation shows that the amplitude of sidebands is related to: (1) the effective HF power; (2) the HF frequency; (3) the temperature of the electrons; (4) the scale height of the ionosphere; and (5) the difference frequency of the two pump waves. Observations of HF sidebands were again made in January, 1986. The 430 MHz incoherent radar provided the ionospheric background measurements. The modulation index theory provides the most complete interpretation of the observations.

Ground-based and Voyager observations in and out of methane and ammonia bands are analyzed to search for longitudinal variations in Saturn's equatorial region. A model with reflecting layer at 2.1 bars, an extended haze to 170 mb and an overlying thin stratospheric haze is adopted. Two sets of data are analyzed, a set of ground based observations covering the 6000-6600 A spectral region and a set of Voyager 1 images obtained with the orange and methane filters. The spectral variations are not consistent with a variation in the height of the reflecting layer. They are modeled by variations in the single scattering albedo of the haze and in the specific abundance of gas in the haze. The ground based spectra, having a spatial resolution of 21,000 km, are consistent with a specific abundance of gas in the haze of 14 $\pm$ 1 km-amagats per mean free path and a haze single scattering albedo in the continuum at 6055 A of 0.990 ($\pm$.006) with a longitudinal variation of $\pm$0.003. The single scattering albedo derived from the ground based observations at 6475 A is 0.993 ($\pm$.006) with a longitudinal variation of $\pm$.003. The Voyager data, having a spatial resolution of $\sim$500 km, are consistent with a specific abundance of gas in the haze between 10b and 24 km-amagats. The larger variation in the specific abundance derived from the Voyager data set is due to the larger relative uncertainty in the intensity (2%) for Voyager as compared to 1% for the ground based spectra. We derive a methane mixing ratio of 2.2 $(\sbsp{-0.2}{+0.8})$ $\times$ 10$\sp{-3},$ representing a C/H ratio which is enhanced by a factor of 2.3 over the solar value. Our estimate of the ammonia mixing ration, 4.5 $\times$ 10$\sp{-4},$ is a lower limit due to our assumption that ammonia exists at its saturation vapor pressure everywhere above the reflecting layer. There is no conclusive evidence that there are longitudinal variations in the structure of the Saturnian atmosphere in the Equatorial Region on a scale greater than 600 km.

A two-dimensional numerical model is created to calculate the electric field and current that flow from a thunderstorm source into the global electrical circuit. The model includes a hemisphere in which the thunderstorm is located, an equalization layer, and a passive magnetic conjugate hemisphere. To maintain the fair weather electric field, the output current from the thunderstorm is allowed to spread out in the ionosphere or flow along the magnetic field lines into the conjugate hemisphere. The vertical current is constant up to $\sim$65 km, decays and is redirected horizontally in the ionosphere. Approximately half of the current that reaches the ionosphere flows along magnetic field lines into the conjugate hemisphere while the rest is spread out in the ionosphere and redirected to the fair weather portion of the storm hemisphere. Our results show that it is important to include a realistic model of the equalization layer to evaluate the role of thunderstorm charging of the global circuit. The mapping of thunderstorm electric fields at middle and subauroral latitudes into the magnetic equatorial plane is studied. The geomagnetic field lines are assumed to be dipolar above $\sim$150 km. The horizontal electric field computed in the ionosphere by our model is of sufficient size and shape for the formation of electron density irregularities in the magnetosphere. The mechanism involves a localized convection of ionization tubes by E x B drift. It is shown that the horizontal range of the electric field disturbance in the ionosphere must be within $\sim$160 km to produce density irregularities necessary for the formation of whistler ducts, well within the predicted size of thunderstorm produced electric fields. Although the electric field strength at ionospheric heights depends sensitively on the conductivity profile, the results presented show that whistler duct formation is possible by thunderstorm generated electric fields.

The climate system exhibits behavior on a wide range of time and spatial scales. Computer models that simulate climate must be faithful to these matters of scale in order to achieve meaningful results. This study involves the consolidation of two models: a model that simulates the seasonal fluctuations of the atmosphere and ocean with a time constant of one day and a continental ice sheet model designed to simulate the behavior of ice sheets over tens of thousands of years. These models are asynchronously coupled in both time and space because computational requirements limit the minimum grid size and time step size. Experiments are presented which examine the long time constant of the continental ice sheet and the fast approach to quasi-equilibrium of the seasonal model. Simulations of present day conditions from the combined model are compared with climate observations and simulations of other climate models.

Considerable progress has been made in the past decades on the life cycle of rapidly deepening winter cyclones. However, little attention has been paid to the roles of mesoscale convective systems (MCSs) in extratropical cyclogenesis that occurs within weak baroclinic environments. In this thesis, the impact of an MCS on the subsequent surface cyclogenesis is investigated, using a 36-h three-dimensional, high-resolution simulations of the famous 10-12 June squall line that occurred during PRE-STORM. The model simulates remarkably well the initiation of the squall line, numerous mesoscale surface pressure perturbations and midlevel circulation structures during the mature stage, and the subsequent surface cyclogenesis after the dissipation of the system. / It is found that the squall line is initiated ahead of a weak surface front with the aid of baroclinic forcing. Once initiated, the squall system is more or less driven by the interaction of convectively generated circulations with the potential unstable environment ahead. The baroclinic forcing only provides a favorable environment for the evolution of the squall system. As the squall system rapidly intensifies and accelerates eastward, it enhances the larger-scale baroclinicity and produces a phase-lag between the pressure and thermal waves so that the baroclinic environment is more favorable for surface cyclogenesis. / To isolate the roles of moist convection in the surface cyclogenesis, a "moist" and a "dry" simulation are compared. It is found that in the absence of moist convection the model could also produce a surface cyclone, but with much weaker intensity, much smaller extent and slower displacement. The effects of moist convection are shown not only to increase the upper-level and decrease the lower-level height of isobaric surfaces, but also to condition the baroclinic environment by increasing the phase lag between the pressure and thermal waves and enhancing the large-scale baroclinicity.

The barotropic interaction in the upper troposphere between the synoptic-scale eddies and the Pacific/North American (PNA) pattern is investigated. The study uses NMC data for the years 1965-89 and examines separately the positive and negative phases of the monthly mean PNA anomalies. Northern Hemisphere maps of the eddy kinetic energy, the barotropic conversion between the eddies and the monthly mean flow and the eddy vorticity forcing are presented for both PNA phases. The eddy vorticity forcing of the PNA anomalies by the eddies is presented in terms of monthly mean geopotential tendencies. / It is found that the geopotential tendencies generated by the synoptic-scale eddies are spatially in phase with the PNA anomalies at the 250 hPa level. The characteristic time scale of the forcing in winter is about 6 to 10 days. The strongest anomaly in the eddy forcing is collocated with the largest amplitude PNA centre over the northern Pacific. / During periods of strong PNA patterns, the synoptic-scale eddy kinetic energy and its transfer into the monthly mean flow is most important in the mean jet exits. The maxima in eddy activity and kinetic energy transfer for the positive PNA phase are situated over the eastern Pacific and are significantly greater than the corresponding maxima for the negative phase, which are located over the central Pacific.

Winter storms produce major problems for society and their varying types of precipitation are often the key factor responsible. The objective of this study is to better understand the formation of winter precipitation types (freezing rain, ice pellets, snow, slush, wet snow and refrozen wet snow) within the varying, and interacting, environmental conditions experienced within many winter storms. A one dimensional cloud model utilizing a double-moment microphysics scheme has been developed to address this issue. Temperature and moisture profiles favorable for the formation of different winter precipitation types were varied in a systematic manner in an environment in which snow is falling continuously from above a temperature inversion. The ensuing precipitation evolved due to atmospheric temperature and moisture variations arising from phase changes such as melting and freezing. Average soundings associated with various surface precipitation types were consequently defined as well as the atmospheric conditions leading to the longest duration of various precipitation events. In addition, the reduction of visibility by precipitation was calculated to determine the conditions leading to the greatest reduction and those leading to its most rapid changes.

A potential vorticity (PV) analysis is performed using a 60-h simulation of a family of mesoscale frontal cyclones which occurred over the Western Atlantic ocean from 0000 UTC 13 to 1200 UTC 15 March 1992. The frontal cyclogenesis events were characterized by an upper-level ring of PV anomalies of stratospheric origin, located on the cyclonic side of the jet stream. The shape of the "PV ring" resembled the tracks of the surface frontal cyclones, implying its role in determining the initiation and the tracks of the surface cyclones. Piecewise PV inversions indicate that low-level baroclinicity contributed most to the intensity of two of the frontal cyclones, followed by low-level latent heating, and upper-level PV anomalies (trough/ridge system). The upper-level PV anomalies contributed most to the depth of a third frontal cyclone. The major frontal cyclone (MFC), underwent explosive deepening (27 hPa in 24 h) and eventually absorbed the low-level circulation of the parent cyclone. The explosive deepening of the MFC can be attributed to the phase-locking between the upper-level and low-level PV anomalies. / A series of full physics sensitivity experiments is performed to test the impact of removing upper-level PV anomalies on the development of the MFC. It is found that the delay in the genesis of the MFC, as well as weakening in its final intensity, is directly related to the proximity of the removed upstream PV anomaly. In another series of "dry" experiment with the removal of upper-level positive PV anomalies, the model fails to generate the MFC when the upstream PV anomaly closest to the pre-MFC is removed. From the results of piecewise PV inversion and sensitivity experiments, low-level thermal advection induced by the flow of the upper-level PV ring appears to initiate and sustain the development of the MFC. (Abstract shortened by UMI.)