The dedicated monitoring web portal for the SWIM (Simulation of RF Wave Interactions with Magnetohydrodynamics) project is now up and running. SWIM is a DoE SciDAC research project charged with integrating disparate physics codes for higher fidelity simulation of magnetically confined plasma. It has physics, math and computer science research components. The web portal http://swim.gat.com:8000, designed and developed by the Data Analysis and Applications Group, communicates with large simulation codes running on computers at ORNL and PPPL and monitors the status of ongoing simulations. The portal also has the capability of real time data visualization and instant messaging-based status notification. The web portal was demonstrated and well received by the SWIM review panel in June.

A closed form for the velocity difference between impurity ions and the main ions in neoclassical theory has been obtained in terms of the pressure and temperature differences. This difference has important diagnostic implications since typically the impurity velocities are measured and the main ion rotation inferred from this. Although the individual velocities cannot be analytically evaluated to the required order in inverse aspect ratio, the difference in a two ion species plasma can be calculated using an approach similar to the one used for bootstrap current. The formula has also been simplified to the case of trace impurity ions, which then is applicable to any number of trace impurities. The resulting difference between two different trace impurity ions in an otherwise pure plasma can be checked against experiments.

The GYRO code has been used for numerical experiments aimed at understanding the fundamental “zonal-flow/drift-wave paradigm” for nonlinear saturation of turbulent transport. The practical conclusion from these numerical experiments is that the drift wave potential fluctuation intensity for each mode wave number should scale roughly like the product of the Geodesic Acoustic Mode (GAM) frequency and the linear growth rate, rather than the square of the linear growth rate as often assumed in models like MMM95. The GLF23 and TGLF nonlinear saturation models are roughly consistent with this rule. The details are in a paper by R. Waltz and C. Holland (UCSD) submitted to Physics of Plasmas.

NIMROD simulations of a 2/1 NTM in DIII-D show that the island saturates at a width ~ 10 cm, qualitatively consistent with experimental observations. The NIMROD simulation was carried out to compare with high-resolution, fast-framing camera images of an m/n = 2/1 Neoclassical Tearing Mode (NTM) in DIII-D. The simulation begins with an EFIT reconstruction obtained in the presence of the saturated NTM. The reconstructed equilibrium thus includes the modified bootstrap current, though it is not evolved self-consistently in NIMROD. The simulation is initialized with 2/1 seed island of ~4 cm at the outer mid-plane; the island then grows for 8 ms and saturates at a width of ~10 cm at the outer mid-plane, consistent with the measured island. A synthetic diagnostic is under development to directly compare the visible bremsstrahlung emission predicted by the simulation with the measured emission.