Grid convergence study in a simulation of LAPD turbulence

Grid convergence of plasma drift turbulence simulations in the Large Plasma Device is studied. When first order upwind advection schemes are used, numerical finite difference diffusion errors cause turbulence statistics, especially saturation levels to vary when grid spacing is changed. The significance and magnitude of the variation is explored, and it is found that use of very fine meshes or higher order advection schemes combined with artificial diffusion and viscosity can lead to grid convergence, while maintaining experimentally consistent saturation levels. It is also found that the details of dissipation range dynamics, such as the level of diffusion, have a significant impact on driven range dynamics due to the lack of a large inertial range in the simulations (copyright 2012 WILEY-VCH Verlag GmbH and Co. KGaA, Weinheim) (orig.)