Quasineutral Particle/Fluid Hybrid Simulation Technique for Whistlers

Summary form only given. We have developed a new simulation scheme for whistlers and other slow EM waves which is fully kinetic and fully nonlinear, works in homogeneous or inhomogeneous situations, is not restricted to a single coherent mode, eliminates the speed-of-light time scale and the electron plasma oscillation time scale, and concentrates simulation resources on the parts of the electron distribution that make kinetic contributions to wave growth. The elements of the scheme are as follows. (1) The plasma is represented as a cold fluid plus a set of simulation particles. The simulation particles can be chosen to represent any part of the electron velocity distribution that is of kinetic interest, e.g. the complete ensemble of energetic electrons, or only a set of resonant electrons. (2) Quasineutrality (QN) is assumed, i.e. div J=0. We emphasize that this does not mean that there are no electrostatic fields, only that the electric field E is determined by QN, rather than by Poisson´s equation. This assumption is appropriate when the frequencies are slow compared to the plasma frequency, and it eliminates plasma oscillations. (3) The displacement current is neglected, since the waves of interest are slow compared to c. This is similar to the Darwin model. However, (1) and (2) allow us to neglect the fall displacement current, not just the solenoidal part, and thereby avoid all of the complications of the Darwin model. (4) Faraday´s and Ampere´s laws, rather than a momentum conservation equation, are used to push the cold fluid velocity. This guarantees QN. (5) The simulation particles are pushed in standard PIC fashion. (6) The magnetic field is determined by Ampere´s law. In the linear regime, the scheme reproduces the quasi-longitudinal dispersion relation for whistlers, and is accurate for propagation angles up to the resonance cone. An extremely stable and accurate predictor-corrector scheme is used, and energy is conserved to great accuracy. The cod- need not resolve spatial scales smaller than the wavelengths of interest, nor time scales shorter than the gyrofrequency. We have used the code to study long-time nonlinear evolution of whistler instabilities; results will be shown for a variety of situations