The Fusion Record — Fusion Energy News ← Home · Knowledge base
Glossary

Gyrokinetics

The theoretical framework for simulating turbulent transport in magnetized plasmas, reducing six-dimensional kinetics to five dimensions by averaging over fast gyromotion.

Reviewed Last reviewed: 9 Aug 2026 · Category: Glossary

From Particle Orbits to Guiding Centers

In a magnetized plasma, charged particles spiral rapidly around magnetic field lines at the cyclotron frequency—typically 108–1011 Hz for ions and electrons in fusion-relevant fields. Resolving every gyration in a kinetic simulation would be computationally prohibitive. Gyrokinetics eliminates this fast timescale by averaging the Vlasov–Maxwell system over the gyrophase angle, reducing the six-dimensional phase space (three spatial, three velocity) to five dimensions: three guiding-center coordinates plus parallel velocity and magnetic moment.1

The Gyrokinetic Ordering

The framework rests on a formal ordering parameter ε ≡ ρi/L, where ρi is the ion Larmor radius and L is the equilibrium gradient scale length. Fluctuations in density, temperature, and potential are assumed to be O(ε) relative to equilibrium values, while their perpendicular wavelengths are comparable to ρi. Frequencies of interest are much lower than the cyclotron frequency—on the order of the diamagnetic drift frequency. This separation of scales is what makes the gyrophase averaging mathematically rigorous and physically justified.2

Why it matters for fusion: Gyrokinetic simulations are the primary tool for predicting turbulent heat and particle losses in tokamaks and stellarators. Codes such as GENE, GS2, GYRO, and GTC underpin the design basis for ITER and next-step devices by quantifying how micro-instabilities—ion temperature gradient (ITG) modes, trapped electron modes (TEM), and electron temperature gradient (ETG) modes—drive cross-field transport that far exceeds collisional predictions.

Modern Gyrokinetic Codes

Two broad numerical approaches dominate. Eulerian (continuum) codes discretize the five-dimensional distribution function on a fixed grid, offering low noise but high memory cost. Particle-in-cell (PIC) codes sample the distribution with marker particles, trading statistical noise for scalability to larger volumes. Both approaches must handle the nonlinear coupling between fluctuating fields and the distribution function that generates turbulent cascades and self-organized structures like zonal flows.3

Global gyrokinetic simulations—retaining profile variation across the full plasma radius—have become feasible on petascale computers, enabling studies of turbulence spreading, avalanche-like transport events, and the interplay between micro- and macro-scale instabilities that shape confinement in burning plasmas.4

Sources

  1. Brizard, A.J. & Hahm, T.S. 'Foundations of nonlinear gyrokinetic theory.' Reviews of Modern Physics 79, 421–468 (2007).
  2. Garbet, X. et al. 'Gyrokinetic simulations of turbulent transport.' Nuclear Fusion 50, 043002 (2010).
  3. Candy, J. & Waltz, R.E. 'An Eulerian gyrokinetic-Maxwell solver.' Journal of Computational Physics 186, 545–581 (2003).
  4. Lin, Z. et al. 'Turbulent transport reduction by zonal flows: Massively parallel simulations.' Science 281, 1835–1837 (1998).

Related