Abstract Details
Abstracts
Author: M.J. Pueschel
Requested Type: Poster
Submitted: 2016-02-12 14:35:34
Co-authors: B.J. Faber, J. Citrin, C.C. Hegna, P.W. Terry, D.R. Hatch
Contact Info:
University of Wisconsin-Madison
1150 University Ave
Madison, WI 53703
USA
Abstract Text:
Gyrokinetic simulations of ion-temperature-gradient-driven modes in the HSX stellarator show the existence of a large number of subdominantly unstable eigenmodes with a high degree of orthogonality. Complex mode structures often make identification across parameter scans infeasible, as the number of unstable modes can be on the order of 100. Both subdominant and linearly stable modes are excited to significant amplitudes in the turbulent state, with the most unstable mode only capturing about a fifth of the nonlinear distribution function. Similarly, employing an orthogonalized set of eigenmodes, the total amplitude contribution of all unstable modes combined reproduces about two thirds of the nonlinear state---this value is slightly lower than that for only the most unstable mode in the tokamak. This affects modeling of transport by quasilinear approaches; such models use mixing-length estimates to predict nonlinear saturated fluxes based on linear simulations, using a single nonlinear run as a scalar gauge. Is is shown here that quasilinear models fail when accounting only for the most unstable eigenmodes. However, when all linearly unstable eigenmodes are included, quasilinear models are able to yield good predictions of ITG transport levels in HSX geometry, making available a powerful tool for efficient stellarator optimization. For this purpose, a sum over unstable modes is required, alotting separate heat flux contributions to any destabilized modes, while making the cumbersome tracking of individual modes superfluous. Successful application of such quasilinear models is not universal, however, as cases of trapped-electron-mode turbulence have been identified where turbulent transport relies partially on nonlinearly excited but linearly stable modes.
Comments: