We have undertaken a study of the low and mid latitude ionospheric electric field pattern, during both magnetospherically quiet and active periods. Our analysis can be conveniently split into two parts. i.In an effort to study the penetration of magnetospheric electric fields to low latitudes, we have compared Jicamarca F-region vertical drifts for 10 radar-observation periods with the auroral boundary index (ABI). The ABI is the latitude of the equatorward edge of the diffuse aurora at local midnight, as estimated from precipitating-electron fluxes measured from DMSP spacecraft. The periods occurred in the interval January 1984 to June 1991 inclusive and each lasted between 2 and 5 days. We focus on periods that occurred in September 1986, March 1990, and June 1991. In the post-midnight sector, where we expect the penetration to be strongest, we found many examples of correlation; specifically, associated with an ionospheric updraft (implying an eastward electric field) is a strong poleward motion of the auroral boundary. However, we also found a significant number of cases where there was little or no correlation. We conclude that there is only mediocre agreement between the observed Sudden Postmidnight Ionospheric Events (SPIEs) and the ABI. These SPIEs have also been compared with other magnetospheric parameters, namely $D\sb{\rm st}$ IMF $B\sb{z}$ and the polar cap potential. $D\sb{\rm st}$ showed significantly better correlation with the SPIEs. We summarize the proposed models for SPIEs and compare their predictions with the data, concluding that no single model can account for all events. While it is clear that some of these SPIEs can be explained in terms of direct penetration of magnetospheric electric fields, we suggest that the remainder may be due to magnetospherically-generated neutral wind effects. ii. We have constructed a model of the low- and mid-latitude potential distribution, applicable for both quiet and active times. We use the Mass-Spectrometer-Incoherent-Scatter (MSIS) model to input the number densities and temperature of the neutral species, and the International reference Ionosphere (IRI) model to input the electron/ion densities and temperatures. As our wind input we use the Horizontal Wind Model (HWM). We find that our model can reproduce the all of the main features of the low latitude ionosphere during quiet times, and supports some of our ideas about magnetospheric penetration during active periods. We use the model to probe the dependency of the low latitude penetration pattern on solar conditions and season and found that the inferred equatorial drifts are relatively insensitive to either. Thus we conclude that ionospheric pre-conditioning is unlikely to play a significant role. On the other hand, the low latitude penetration pattern is strongly dependent on the assumed poleward boundary.

A steady state flow/field model is applied to the direct interaction of the solar wind with the Martian ionosphere. We have coupled observational data with the kinetic theory plasma equations within a single self-consistent framework. We do not imply that the interaction is purely ionospheric, but rather we show that a self-consistent, direct ionospheric interaction is feasible. The model demonstrates how the dynamic solar wind pressure is transferred to the ultimate obstacle, the planet itself, without the requirement of an intrinsic Martian magnetic field. The results of the model are in agreement with Viking observations, and hence make possible predictions of quantities as yet unmeasured within the Martian ionosphere. We predict that the magnetic field configuration of the Martian ionosphere may be similar to the "Venusian dip" structure observed by Pioneer Venus. (Abstract shortened with permission of author.)

This thesis describes the background behind and early results from an instrument designed to measure atmospheric electrical currents. Preliminary data presented herein (including the marked similarity in the signals from the two arrays, the noticeable Carnegie curve, and the correlation between the measured electric field and current at the same site) demonstrate the proper functioning of the instrument. A description of global circuit theory and Antarctic climate illuminate the deployment decisions. Also presented are a preliminary study suggesting that the instrument is affected by the presence of an electrode layer, and some of the mathematical and theoretical relationships required to determine physical variables from the raw data.

The exchange of moisture and heat between the atmosphere and the Earth's surface fundamentally affect the dynamics and thermodynamics of the climate system. In order to trace moisture flow through the climate system and examine its impact on climate, a hydrologic cycle and a land energy balance have been developed and incorporated into a coupled climate-thermodynamic sea ice (CCSI) model. The expanded CCSI model has been tested by comparing computed climate parameters with available observations and GCM modeling results. In general, the expanded model does a good job in simulating the large scale features of the atmospheric circulation and precipitation in both space and time. The expanded model has been used to examine the possibility that increased levels of CO$\sb2$ in the atmosphere may induce the growth of Northern Hemisphere ice sheets. Results of the study indicate that if summer ice albedo is high enough, and there is some mechanism for initially maintaining ice through the summer season, then it may be possible to have ice sheet growth under the conditions CO$\sb2$ induced warming, mainly the result of decreased summer ice melt in response to the higher land ice albedo, and not an increase in precipitation. The expanded model has also been used to examine the impact of Milankovitch solar radiation variations on the climate system, to study the mechanisms that produce glacial-interglacial cycles, especially with respect to the initiation of ice sheets. The results show the Milankovitch solar radiation variations affect the climate system most in the polar regions with the mean annual surface air temperature varying directly in response to changes in the annually averaged incoming solar radiation. However, the seasonal variations in the surface air temperatures are much more complex with large magnitude variations for brief times during the year. The study indicates that ice sheets may start to grow under the conditions of low insolation that occurred at 25, 70, and 115 kyr BP and a land ice minimum albedo of 0.53, with the largest growth rate at 115 kyr BP, approximately when the current 100 kyr cycle began as observed in the geological record.

A model is presented which simulates the behavior of superthermal ions previously reported in the dayside ionosphere of Venus. The model considers effects of E $\times$ B and gradient drifts, charge exchange and collisions with the ambient neutral atmosphere and the possible effects of a wave-particle (anomalous) scattering process. Results indicate that scattering processes are required if superthermal ions are the explanation for the observed "missing pressure" component in the dayside Venus ionosphere. The scattering scale length required to match the "missing pressure" distribution is similar to the scale length previously predicted for growth of a lower hybrid beam instability.

Due to their formation at about the same time in the same region of the early solar nebula, it is reasonable to assume that the primitive atmospheres of Earth, Mars and Venus were similar and that present-day differences have arisen as a result of their differing masses and incident solar fluxes. Using a radiative-convective model, we determine maximum and minimum carbon dioxide levels for the early atmospheres which are consistent with this assumption and with climatic conditions thought to have existed on the three terrestrial planets 4.0 billion years ago. Rather than employ the cloud-free atmosphere approach of earlier studies, we include an interactive water vapor transport and cloud formation scheme in the model. Due to uncertainties about the direction of cloud cover feedback, we treat cloud cover as fixed. For most cases examined, we set the cloud cover at 50%, but the effect of varying cloud cover is also explored.

The magnitude, spatial extent, growth, and seasonal and diurnal behaviors of the urban heat island of Houston, Texas are characterized using both in situ air temperature and remotely sensed surface temperature data. Between 1990 and 2000, the air temperature heat island of Houston had an average magnitude of 1.25 K at night but was largely absent during the day. This behavior is reflected in a survey of extreme temperature events, which reveals a dramatic increase in the number of extremely warm nights relative to the surrounding rural areas. Thermal satellite imagery acquired between 1985 and 2001 demonstrate a surface temperature heat island of approximately 3 K at night and up to 10 K during the day. Climatological analysis reveals an inverse dependence of air temperature heat island magnitude on rural temperature. Conversely, daytime surface temperature heat islands grow with rural temperature, while nighttime surface temperature heat islands show no relationship to rural temperature. Examination of temperature maps reveals an urban heat island area of 1200 km2 at night and 2100 km2 during the day. Comparison of satellite imagery taken twelve years apart exhibits a growth in the nighttime heat island of 0.8 K in magnitude and 650 km2 in area. High-resolution temperature data are also examined and show an urban temperature dependence on population density.

Once a day, a flash of gamma-rays erupts somewhere in space and is detected by an international fleet of satellites. Since their first detection over a quarter century ago, these gamma-ray bursts have puzzled researchers who could not determine their distance, emission mechanism, or progenitor. Much of this confusion arose as theorists attempted to create a single model to explain what we now believe are at least two, and probably more, populations of gamma-ray transients. Within the past two years, thanks largely to the Dutch-Italian satellite BeppoSAX, astronomers discovered that bursts have multiwavelength fading afterglows. This helped them determine that most gamma-ray bursts are from distant galaxies. However, it did not answer the questions regarding the emission mechanism or the progenitor. We place constraints on the emission mechanism by closely examining the spectral evolution of gamma-ray bursts observed by the American instrument BATSE. From a sample of 41 distinct pulses in 26 bright gamma-ray bursts, we have determined that the pulses appear to be radiatively cooling. We also studied the evolving spectral shape in 79 bursts. In particular, we found that both the range and evolution of the spectral index below the spectral break conflict with the predictions of a popular synchrotron shock model. They instead suggest inverse Comptonization in a hybrid thermal plus nonthermal plasma as the emission mechanism. With our Monte Carlo codes, we have begun the generation of a library of inverse Compton spectra. Using them, we have made preliminary fits to two bursts with prompt multiwavelength data. The characteristic "terrace-shaped" Compton spectrum is evident in both using BATSE data alone. This shape appears to be confirmed for the January 1, 1997 burst using BeppoSAX X-ray data and for the January 23, 1999 burst using optical data from the ground-based robotic telescope, ROTSE. Both bursts appear to be enshrouded in a material with a high initial Thomson scattering depth tT≳20 . Fitting with a larger, more organized library of Monte Carlo simulations will be required before precise limits can be placed on physical parameters such as the energies, masses, and densities of bursters.

Review and analysis of atomic collision data (such as collision cross sections and ion-neutral reaction rates) used in the upper atmospheric modeling for Earth and other planets are presented. This work gives the first complete compendium of the electron impact cross sections and ion neutral reaction rates used in aeronomic studies. This data are presented in an easy-to-use form, allowing atmospheric modelers to analyze what data were used in each existing model and permitting a choice of the best or most recent data to use in future models. An original windows-based program provides a straightforward means for viewing different upper atmosphere models altogether with the reactions and reaction rates used in each model.

Mars Global Surveyor's magnetometer/electron reflectometer (MAG/ER) experiment has returned over eight years of observations of low frequency plasma waves produced in the interaction of the solar wind with the Martian ionosphere. Using the MAG/ER data, I identify the properties and physical origins of the waves in the magnetosheath, magnetic pileup region, and ionosphere. I find that the waves in the dayside magnetosheath are predominately mirror mode instabilities produced by plasma temperature anisotropies arising from the draping of the solar wind magnetic field around the ionosphere. The nightside magnetosheath shows evidence for resonant ion instabilities arising from the interaction of the solar wind plasma with the ionospheric plasma. These waves are therefore an indirect observation of ongoing atmospheric loss at Mars. During the large solar storm of October 2003, dramatic changes were observed in the plasma waves present; even the normally placid tail region showed signs of significant wave activity. Coherent oscillations are observed in the ionosphere and are presumably driven by solar wind fluctuations or are associated with field line resonances along crustal fields.