Particle-in-cell Simulations of Decaying Plasma Turbulence: Linear Instabilities versus Nonlinear Processes in 3D and 2.5D Approximations

Abstract

Abstract Particle-in-cell (PIC) simulations are used to examine the decay of strongly intermittent Alfvénic turbulence in a collisionless, homogeneous, and magnetized plasma. Results from three computational models are compared in simulations with similar plasma parameters and dimensions of approximately 100 d i, where d i is the ion inertial length. Each model utilizes three-dimensional velocities, but spatial variations differ: the 2.5D perpendicular PIC model uses two-dimensional spatial variations with the background magnetic field B o perpendicular to the simulation plane, the 2.5D parallel PIC model uses two-dimensional spatial variations with B o in the simulation plane, and the 3D model includes spatial variations in full three-dimensional space. Results from the three models are compared using plots of the joint probability distribution functions (PDFs) of maximum local linear instability growth rates versus the maximum local nonlinear frequencies. All results agree with previous demonstrations that linear growth rates are generally slower than the nonlinear frequencies of the turbulence at kd i = 1.0. However, it is the 3D PIC joint PDFs that most closely resemble joint PDFs recently observed in space plasmas because the 3D PDFs capture both the linear and nonlinear plasma processes, whereas the 2.5D parallel PIC runs do not represent the nonlinear turbulence processes and the 2.5D perpendicular PIC computations do not well represent the consequences of microinstabilities. These results suggest that 3D simulations are needed to properly capture important features of both microinstabilities and nonlinear turbulence.