Hybrid modeling of plasma and applications to fusion and space physics

Obtaining reasonable solutions to the nonlinear equations is crucial to the understanding of the behavior of plasmas. With the advent of high speed computers, computer modeling of plasmas has moved into the front row of the tools used in research of their nonlinear plasma dynamics. There are roughly speaking two types of plasma models, particle models and fluid models. Particle models try to emulate nature by following the motion of a large number of charged particles in their self consistent electromagnetic fields. Fluid models on the other hand use macroscopic fluid equations to model the plasma. MHD models are typical of this type. Particle models in general require larger memory for the computer due to the massive amount of data associated with the particles' kinematical variables. Particle models are generally limited to studying small regions of plasma for relatively short time intervals. Fluid models are better fit to handle large scales and long times; i.e., quite often the complete plasma involved in an experiment. The drawback of the fluid models however is that, they miss the physical phenomenon taking place at the microscale and these phenomenon can influence the properties of fluid. Another approach is to start with fluid models and incorporate more physics. Such models are referred to as hybrid models. In this thesis, two such models are discussed. They are then applied to two problems; the first is a simulation of the artificial comet generated by the AMPTE experiment; the second is the production of enhanced noise in fusion plasmas by injected energetic ions or by fusion reaction products. In both cases the models demonstrate qualitative agreement with the experimental observations