Experimental constraints on transport from dimensionless parameter scaling studies

The scalings of heat transport with safety factor $\mathrm{(q)}$ , normalized collisionality $\mathrm{(\nu )}$ , plasma beta $\mathrm{(\beta )}$ , and relative gyroradius ${\mathrm{(\rho }}_{*})$ have been measured on the DIII-D tokamak [Fusion Technol. 8, 441 (1985)]. The measured ${\rho }_{*},$ $\beta$ and $\nu$ scalings of heat transport indicate that $\mathrm{E\times B}$ transport from drift wave turbulence is a plausible basis for anomalous transport. For high confinement (H) mode plasmas where the safety factor was varied at fixed magnetic shear, the effective (or one-fluid) thermal diffusivity was found to scale like ${\chi }_{\mathrm{eff}}{\mathrm{\propto q}}^{\text{2.3±0.64}}$ , with the ion and electron fluids having the same $q$ scaling to within the experimental errors except near the plasma edge. The scaling of the thermal confinement time with safety factor was in good agreement with this local transport dependence, ${\tau }_{\mathrm{th}}{\mathrm{\propto q}}^{\text{−2.42±0.31}}$ ; however, when the magnetic shear was allowed to vary to keep $q_0$ fixed during the (edge) safety factor scan, a weaker global dependence was observed, ${\tau }_{\mathrm{th}}{\mathrm{\propto q}}_{95}^{\text{−1.43±0.23}}$ . This weaker dependence was mainly due to the change in the local value of $q$ between the two types of scans. The combined ${\rho }_{*},$ $\beta$ , $\nu$ and $q$ scalings of heat transport for H-mode plasmas on DIII-D reproduce the empirical confinement scaling using physical (dimensional) parameters with the exception of weaker power degradation.