Theory Weekly Highlights for November 2020

November 25, 2020

In order to understand the dependence of turbulence on flux surface shape, a study of the 3D spectrum of potential fluctuations from gyrokinetic turbulence computed with the CGYRO code was undertaken. The findings have been accepted for publication in Plasma Phys. Control. Fusion It was found that the poloidal variation of the peak fluctuation intensity could be modeled by geometric factors present in the metric coefficients of the poloidal and radial wavenumbers. The elongation and Shafranov shift were varied in order to determine the geometric dependence of the width of the radial wavenumber spectrum and the peak amplitude of the potential spectrum as a function of the poloidal wavenumber. The width of the zonal (axisymmetric) potential fluctuation spectrum was found to set a minimum width for the rest of the spectrum controlling the net fluxes. A model of the 3D spectrum that gives a highly accurate quasi-linear flux model has been incorporated as an option in the TGLF transport model.

November 20, 2020

A GA theory team led by Emily Belli, along with Jeff Candy, Gary Staebler, and George Fann from ORNL, has received a 2021 Innovative and Novel Computational Impact on Theory and Experiment (INCITE) Award for their proposal “Multi-Ion Turbulence in Burning Plasma Experiments“. The award allocation consists of 450,000 node-hours on Summit, the fastest supercomputer in the US, based at the Oak Ridge Leadership Computing Facility. The team will use the CGYRO code to perform “capability computing” simulations to model energy and particle losses due to turbulent transport in burning plasma scenarios and develop an understanding of how complex multi-ion interactions between D and T fuel ions, fusion helium ash, wall impurities (beryllium and tungsten), and electrons affect plasma performance. The CGYRO algorithm has been highly optimized for scalability on advanced HPC systems like Summit that use multicore and GPU-accelerator hardware. These simulations will make optimal use of the leadership-scale multi-species and multiscale plasma turbulence simulation capabilities of CGYRO in understanding burning plasmas in preparation for ITER.

November 13, 2020

Three invited talks were presented (remotely) at the APS Division of Plasma Physics conference this week:

Emily Belli presented an invited talk on “Strong Reversal of Simple Isotope Scaling Laws in Tokamak Edge Turbulence”

Valerie Izzo presented an invited talk on “MHD modeling of dispersive shell-pellet injection as an alternative disruption-mitigation technique”

Xiang Jian presented an invited talk on “Destabilization of High-Field-Side Micro-Instabilities by Large Shafranov Shift in Present and Future Devices”

Emily Belli’s research was also described in an APS press release “Turbulent Relationship: Understanding the Mysterious Hydrogen 'Isotope Effect' in Fusion Experiments” (see

In addition, Valerie Izzo’s research, together with that of Robert Wilcox, was described in an APS press release:
“Taming Fusion Plasmas with Ice Cubes and Diamond Shells” (see

November 06, 2020

A big data approach to validating tokamak transport models will be reported by Tom Neiser (ORISE) at the APS-DPP meeting next week. A database of 80000 time and space points was constructed from 2500 randomly selected plasma discharges in DIII-D. A set of good data selection filters was then applied to curate the dataset. The quasi-linear transport model TGLF was then successfully tested against this curated dataset. Machine learning tools are being used to identify any promising areas of improvement for the model.

These highlights are reports of research work in progress and are accordingly subject to change or modification


Theory Weekly Highlights for December 2020

December 18, 2020

The Stability, Transport, Equilibrium, and Pedestal (STEP) workflow, implemented in the OMFIT integrated-modeling framework, has been validated against a DIII-D negative-central-shear (NCS) plasma, an advanced-tokamak scenario that improves MHD stability by maintaining a safety factor greater than two throughout the plasma. By coupling the Current, HEating, and Fueling (CHEF) module for sources, ONETWO for current evolution, EFIT for equilibrium calculation, and TGYRO (TGLF + NEO with EPED neural networks) for core transport, STEP was able to accurately recreate the equilibrium and kinetic profiles seen in experiment. While previously validated against standard H-modes, STEP required several upgrades to reproduce this more-exotic scenario. A current source driven by toroidal-field ramp was added to CHEF, allowing STEP to incorporate this significant source of off-axis current used in the experiment. In addition, an ad-hoc fast-ion diffusion model was added to FREYA neutral-beam code heating and current-drive calculations. Initial simulations with DCON and GATO in STEP have been carried out for this plasma, looking at the ideal magnetohydrodynamic stability of the scenario. Future work will seek to vary the neutral-beam power in STEP to calculate the observed ideal stability limit in NCS plasmas, with the goal of creating a theory-based predictive stability tool that can be used in the design of future tokamak experiments and reactors.

These highlights are reports of research work in progress and are accordingly subject to change or modification


Theory Weekly Highlights for January 2021

January 15, 2021

An integrated modeling workflow capable of finding the steady-state plasma solution with self-consistent core transport, pedestal structure, current profile, and plasma equilibrium physics has been developed and successfully tested against a DIII-D discharge. Key features of the simulation were the self-consistent inclusion of impurity transport and the use of machine learning accelerated models for both the pedestal structure and the turbulent transport physics. The coupled workflow is implemented within a new Stability Transport Equilibrium Pedestal (STEP) physics module in the One Modeling Framework for Integrated Tasks (OMFIT) framework. STEP makes use of the ITER integrated modeling and analysis suite data structure for exchanging data among physics codes, and such technical advance was enabled by the development of a new numerical library named the Ordered Multidimensional Arrays Structure (OMAS). Simulations of an ITER baseline scenario within STEP show that although D-T fuel dilution in the core raises the core plasma temperature, this positive effect is offset by a decrease in the available fuel, a decrease in pedestal pressure, and increased radiation. The work has recently appeared in Meneghini, et al., 2021 Nuclear Fusion 61 026006.

January 08, 2021

Simulations of negative triangularity (δ < 0) plasmas predict relatively good confinement, (H-mode scaling factor H98y2 ~ 1 ) with only a small H-mode edge pedestal, consistent with DIII-D experimental observations. Profile predictions for this scenario using the TGYRO transport manager show good agreement with experiments using either the TGLF-SAT0 and TGLF-SAT1 models for prediction of nonlinear saturation of turbulent transport. Negative-δ is found to reduce the predicted pedestal pressure by a factor of 2.5. TGLF predicts that turbulence is reduced across all wave numbers, and that an increase in the electron temperature gradient scale length relates to a reduction in particle transport. In agreement with experiments, core-pedestal STEP (Stability, Transport, Equilibrium, Pedestal) modeling finds that confinement increases as δ is reduced below zero. This is attributed to the increase of electron temperature gradient scale length reducing the particle transport.

These highlights are reports of research work in progress and are accordingly subject to change or modification