(TH/2-6Rb) Long Time Simulations of Microturbulence in Fusion Plasmas

W.W. Lee1), S. Ethier1), T. G. Jenkins1), W. X. Wang1), J. L. V. Lewandowski1), G. Rewoldt1), W. M. Tang1), S. E. Parker2), Y. Chen2)
1) Princeton Plasma Physics Laboratory, Princeton, NJ, United States of America
2) Center for Integrated Plasma Studies, University of Colorado, Boulder, CO, United States of America

Abstract.  Remarkable success has been achieved in the area of kinetic simulations of turbulence transport in tokamaks using the modern massively parallel computers. Here, we will report the use of the δf global toroidal gyrokinetic particle simulation code (GTC) for studying the long time behavior of microturbulence for fusion plasmas. Establishing the capability of this type is extremely important for making meaningful comparisons between the simulation observations and actual experimental measurements. One of our findings is that the often-neglected velocity space nonlinearity associated with parallel acceleration has a profound influence on the production of zonal flow and the temporal evolution of the ITG turbulence. This parallel acceleration term, which must be taken into account to ensure the energy conservation, is neglected in most of the fusion turbulence codes. Secondly, to address the questions about the intrinsic particle noise in PIC codes, we will present our recent findings by extending the Fluctuation-Dissipation Theorem to a nonlinearly saturated system arising from drift instabilities. Based on this ``first principles” approach, it is demonstrated that the discrete particle noise, by using an insufficient number of particles, will always enhance the steady state particle (and thermal) flux in accordance with the entropy conservation property. This is in disagreement with the conclusion made by a recent study, which claims that discrete particle noise can suppress steady state thermal flux. In addressing the issues of special importance for burning plasmas such as ITER, we have also carried out studies of the influence of electromagnetic effects on turbulent transport using GEM, a δf particle code for turbulence studies with kinetic electron dynamics and electromagnetic perturbations. With the recent extension to handle general toroidal equilibrium magnetic field configurations to enable realistic applications to actual experimental scenarios, we have used GEM for high-β NSTX plasmas for studying kinetic ballooning modes (KBM) and toroidicity-induced-Alfvén-eigenmodes (TAE). Details will be reported.
*  Work supported by the DoE Contract DE-AC02-76-CHO-3073 and SciDAC GPS Center.

Full paper and slides available (PDF)