GA in partnership with MIT Plasma Science and Fusion Center (PSFC) and Lawrence Berkeley National Laboratory (LBNL) has begun a three-year project to track data provenance. This work is general in nature so as to benefit all Office of Science programs, but is being field tested on magnetic fusion energy research - specifically DIII-D and C-Mod data. Data provenance tracking is for both experimental and simulation data. The approach is to create a general Application Programming Interface (API) that can be called from most any language (Python script, IDL, MDSplus, etc.) allowing easy workflow instrumentation and therefore automatic data provenance tracking. An initial capability including a web site for searching and displaying results is operating to allow early user feedback.
The NEO kinetic neoclassical code has been used to study the accuracy of the Sauter model for the bootstrap current and the new formula by Koh and Chang et al. The Koh bootstrap current formula is a modification of the Sauter model and is designed to bring closer agreement between the XGC0 drift-kinetic neoclassical edge transport code and the analytic bootstrap expression in the pedestal region. For typical DIII-D plasmas, the Koh modification is negligible. However, for NSTX plasmas in the pedestal, the Koh formula predicts a much larger magnitude of the bootstrap current than the Sauter model. The NEO results generally follow the Sauter model, with a slightly lower value, and thus do not agree with the Koh modification formula. A simple example in s-alpha geometry illustrates that there is an anomaly in the Koh formula at low aspect ratio related to the terms proportional to the temperature gradient. In this case, we set the ion neoclassical driving gradients to zero and retain only a moderate electron temperature gradient, which eliminates any caveats regarding steep gradients. At large aspect ratio, NEO, Sauter, and Koh agree reasonably well across a wide range of collisionality. But, as the aspect ratio decreases, the Koh formula develops a large spike near the region where the electron collisionality approaches unity, which appears unphysical. In this region, NEO qualitatively agrees with Sauter.
Three-dimensional (3D) tokamak equilibria have been calculated using the 3D two-fluid MHD code M3D-C1 for a variety of purposes, including the evaluation of proposed non-axisymmetric coils on ITER and DIII-D, comparison of perturbed edge quantities with experiments on DIII-D, the prediction of edge displacements on ITER, and 3D transport. These calculations are now being used to understand RMP ELM suppression and mode locking on DIII-D. Additionally, linear and nonlinear M3D-C1 calculations played a significant role in the code-benchmarking activity that comprised the 2012 Fusion Energy Science Theory Milestone on the effects of non-axisymmetric magnetic fields in tokamak equilibria, coordinated by Allan Reiman of PPPL. These calculations provided a basis for understanding under which circumstances the plasma response can be accurately described using linear perturbation theory.
As reported in a previous highlight, the code ALPHA has been programmed to project the ITER profiles for the fusion alpha classical slowing down density, equivalent Maxwellian alpha temperature, and cross-over energy all from the EPED1 predicted pedestal beta and the TGLF predicted plasma ion temperature and density peaking factors (over the pedestal values). The ALPHA code has been extended to predict the transported fusion alpha density profiles relative to the classical slowing down density profiles. The simple 1D transport model pins the alpha density gradient to the low-n Alfven eigenmode (AE) threshold gradient in the mid-core where the AE modes are stable with the GYRO fitted Angioni model for the high-n micro-turbulent Alpha diffusion. Very early results suggests that the micro-turbulent alpha diffusion near the edge may be more significant (for wall damage) than the reduction in Alpha plasma heating by unstable AE modes. ALPHA will be used for the forthcoming GYRO energetic particle (EP) Alfven mode ITER stability studies (DoE 2014 EP Milestone).
The Data Analysis and Applications Group is currently working on updating the IDL EFITLoader tool in order to provide greater flexibility of EFIT equilibrium reconstructions storage in the MDSPlus database. The focus of these changes is on storage of EFIT runs that contain multiple SNAP input files and/or EQDSK equilibrium output files. This will allow users to better understand the EQDSK data stored in a particular MDSPlus tree, the SNAP files used to create that EQDSK data, and assist in reproducing the run where desired.
These highlights are reports of research work in progress and are accordingly subject to change or modification