Simulating Plasma Turbulence in Tokamaks

The development of magnetic fusion energy is a continuing national priority. Present day toroidally confined plasma experiments (tokamaks) are producing fusion power comparable to the input power needed to heat the plasma (Pfusion ≈ 0.3Pin). One of the basic physics problems is to understand turbulent transport phenomena, which cause particles and energy to escape the confining magnetic fields. Plasma turbulence is challenging from a theoretical point of view because of the nonlinearity and high dimensionality of the governing equations. Experimental measurements in the core region of a tokamak are limited by the extremely high temperatures, O(10) Kelvin, of the confined plasma. The high levels of both theoretical and experimental difficulty highlight the potentially important role of numerical simulations in developing a predictive model for turbulent transport. Such a model would dramatically reduce uncertainties in tokamak design and could lead to enhanced operating regimes, both of which would reduce the size and expense of a fusion reactor. An understanding of turbulent transport and the exploration of modes of operation that suppress turbulence are central goals of the numerical tokamak project, one of the Grand Challenges of the national HPCC program. The process of modeling entails three main steps: (1) the development of a simplified model of tokamaks that encompasses the essential physics of the relevant instabilities, (2) the creation of numerical algorithms for solving the governing equations, and (3) implementation of these methods on massively parallel architectures. This third step is necessary if we are to achieve simulations of sufficient size and resolution to explain the trends seen in