Comparing Gyrokinetic Simulation Results and Linear Gyrofluid Closures to Develop Robust Closures*

Gyrofluid models are attractive because they provide a conceptually simpler and computationally efficient alternative to gyrokinetic models. They rely on a moment closure, which approximates the highest order fluid moment as a function of the lower order moments. Conventional gyrofluid models use linear closures designed to match the plasma dispersion function and can produce linear physics that closely matches gyrokinetics. However, these closures break down in the presence of turbulence, where the nonlinearity strongly modifies the kinetic physics. In order to develop a better understanding of the effects of nonlinear phenomena, we analyze a reduced gyrokinetic model. Detailed comparisons are made between the phase mixing in (1) the linear kinetic system, (2) the linear system closed with a Hammett-Perkins-like closure, and (3) the nonlinear turbulent system modeled by the DNA code. The closure predicts growth rates of the ITG mode that closely agree with the full linear kinetic system. However, the linear physics agrees with the dynamics in the nonlinear turbulent system only in very small regions of wavenumber-space. These discrepancies are analyzed in detail with the aim of formulating gyrofluid closures that are robust even in the presence of turbulence.