Blue jets and gigantic jets are transient luminous events in the middle atmosphere that form when conventional lightning leaders escape upward from thundercloud tops and propagate toward the lower ionosphere. These events are believed to be initiated by 'classic' parent lightning discharges, when they escape upward from cloud tops. The present study builds upon a previously introduced lightning model that combines the hypotheses of equipotentiality and overall charge neutrality of the lightning channel with the fractal approach allowing to describe the stochasticity and branching of the discharge.

The modeling indicates that blue jets occur as a result of electrical breakdown between the upper storm charge and screening charge attracted to the cloud top; they are predicted to occur 5–10 s or less after a cloud-to-ground or intracloud discharge produces a sudden charge imbalance in the storm. A new observation is also presented of an upward discharge that supports this basic mechanism. Gigantic jets are indicated to begin as a normal intracloud discharge between dominant midlevel charge and a screening-depleted upper level charge that continues to propagate out the top of the storm. Observational support for this mechanism comes from similarity with 'bolt-from-the-blue' discharges and from data on the polarity of gigantic jets. Upward discharges are analogous to cloud-to-ground lightning and their explanation provides a unifying view of how lightning escapes from a thundercloud.

A two-dimensional axisymmetric model of charge relaxation in the conducting atmosphere is developed. It is used in conjunction with the lightning model to demonstrate how realistic cloud electrodynamics leads to the development of blue and gigantic jets. This model accounts for the time-dependent conduction currents and screening charges formed under the influence of the thundercloud charge sources. Particular attention is given to numerical modeling of the screening charges near the cloud boundaries. The results demonstrate the important role of the screening charges in local enhancement of the electric field and/or reduction of net charge in the upper levels of the thundercloud. This model shows that the accumulation of screening charges near the thundercloud top produces a charge configuration leading to the initiation of blue jets, and the effective mixing of these charges with the upper thundercloud charge may lead to the formation of gigantic jets.

We develop a model of the streamer-to-spark transition to study this transition from cold, weakly ionized plasma to thermalized spark at various altitudes (or equivalently, ambient air densities) in the Earth atmosphere. The model is a fully one-dimensional (1-D) axisymmetric, axially invariant thermodynamics model coupled to a zero-dimensional (0-D) chemical kinetics scheme. In this dissertation, the model is applied to study the scaling properties of air heating in streamer channels under conditions of constant electric field. The model results on characteristic heating times τ_br appear to be in excellent agreement with the available laboratory measurements conducted in short discharge gaps at ground and near-ground pressures. The results demonstrate a significant acceleration of the heating at lower air densities, with effective heating times appearing to scale closer to 1/ N than to 1/N² predicted on the basis of simple similarity laws for Joule heating, where N is the ambient air density. This acceleration is attributed to strong reduction in electron losses owing to three-body attachment and electron–ion recombination with reduction of air pressure. The results also indicate that at low ambient air densities, the channel conductivity and the air temperature increase very rapidly in comparison with the gas dynamic expansion time (i.e., τ _br≤r_s/c_s, where r_s is the streamer channel radius and c_s is speed of sound). Thus both constant-density and constant-pressure approximations to channel dynamics commonly used in previous studies at ground pressure lead to nearly identical streamer-to-spark transition times. (Abstract shortened by UMI.)