Preliminary EFIT and TRANSP analysis of the NBI fast ion pressure profile for DIII-D discharge 146102 is in good agreement with the Alfven eigenmode critical gradient model developed for the ALPHA energetic particle density transport code. This discharge had high minimum safety factor with on-axis NBI. Gyrokinetic simulations with GYRO provided the critical gradient at multiple radii using a beam-like slowing down distribution. While net losses to the edge are small, ALPHA predicts about half the birth fast ions were lost from the central core r/a < 0.5.

Transport prediction with TGLF+NEO of DIII-D high poloidal beta discharges found a significant shortfall in the electron energy transport in the core plasma. Investigation of the possible instabilities that could cause the electron energy transport have shown that increasing the saturated turbulence level for high wavenumber electron temperature gradient (ETG) modes used in TGLF can successfully predict the electron temperature profile for several cases. However, the ion momentum transport is also under-predicted by TGLF, and ETG modes do not produce ion momentum transport because the ions do not respond at the high-wavenumbers where ETG modes are unstable. In order for ETG modes to produce ion momentum transport there would have to be a coupling of ETG modes to low wavenumber ion fluctuations. Just such a multi-scale interaction has been observed in non-linear gyrokinetic turbulence simulations of L-mode edge plasma conditions. This new study with TGLF suggests the same multi-scale physics could be important over the whole core of high poloidal beta plasmas.

An OMFIT workflow has been developed to enable profile prediction using core transport coupled to a dynamic pedestal model. The approach exploits the time-scale separation between plasma equilibrium, transport, and current evolution. Self-consistency is achieved by iterating between the core solution predicted by TGYRO, and the pedestal structure predicted by EPED. A self-consistent equilibrium is calculated by EFIT, while ONETWO evolves the current profile and calculates the particle and energy sources. The plasma is subdivided into 4 radial subdomains to reflect the different physics mechanisms governing profile evolution across the plasma radius. Near the magnetic axis (rho<0.3), the drift ordering breaks down, so the inverse profile scale-length is taken to vary linearly from zero to the innermost TGYRO radius. Similarly, the inverse scale-length is taken to vary linearly between the predictions of the pedestal and core regions (0.8<rho<0.9). The only inputs are the initial plasma equilibrium, the rotation profile, the electron density at the top of the pedestal, the input power, and an initial guess for the normalized plasma beta. Initial validation with a DIII-D ITER baseline scenario discharge shows that the experimental profiles and normalized plasma beta can be predicted with high fidelity, even when the initial guess for the normalized plasma beta is 75% of the experimental condition.

M3D-C1 modeling of AUG discharges in which n=2 fields were applied has found that different n=2 spectra couple to different modes present in the response. In particular, a mode having a strong core response was found to be excited when the RMP coil rows had a -90 degree relative phasing, and the strongest edge kinking response was found at a +90 degree phasing. In the experiment, ELM suppression was obtained at the +90 degree phasing, but not at the -90 degree phasing. This is consistent with published analysis of n=2 ELM suppression in DIII-D, which found ELM suppression to be correlated with the modeled edge kink response. In single-fluid calculations, an edge tearing mode with the same phasing dependence as the core mode was also found to be excited, but this tearing response was eliminated when the diamagnetic flows were included in the rotation profile. This implies that two-fluid effects may be important for understanding the plasma response in the steep-gradient region. Similar analysis for recent and upcoming DIII-D experiments is planned.