2. COMPARISON OF THEORETICAL AND EXPERIMENTAL HEAT DIFFUSIVITIES IN THE DIII-D EDGE PLASMA

Predictions of theoretical models for ion and electron heat diffusivity have been compared against experimentally inferred values of the heat diffusivity profile in the edge plasma of two H-mode and one L-mode discharge in DIII-D (J. Luxon, Nucl. Fusion, 42, 614 (2002)). Various widely used theoretical models based on neoclassical, ion temperature gradient modes, drift Alfven modes and radiative thermal instability modes for ion transport, and based on paleoclassical, electron temperature gradient modes, trapped electron modes, and drift resistive ballooning modes for electron transport were investigated. A. Introduction The structure of the density and temperature profiles in the edge of tokamak plasmas has long been an area of intense research, at least in part because of the apparent correlation of this structure to global plasma performance. Essential to an understanding of the structure in the edge density and temperature profiles, in the absence or in between edge-localized-modes (ELMs), is an understanding of the underlying transport mechanisms. A methodology for inferring the underlying heat diffusivities from measurements of temperature and density profiles in the plasma edge, which takes into account convection, atomic physics and radiation cooling, ion-electron energy exchange and other edge phenomena, has recently been developed and applied to several different types of DIII-D1 discharges 2-5. While some comparisons with theoretical formulas have been included in this previous work, the emphasis was on the development of accurate fits of the measured data for use in the inference of experimental heat diffusivity profiles and the accurate calculation of heat and particle fluxes to be used in these inferences. The purpose of this paper is to report a comparison of several theoretical predictions of heat diffusivities with the experimentally inferred heat diffusivities, primarily to gain insight as to the more likely transport mechanisms in the plasma edge, and secondarily to compare some currently used transport models with experiment. To this end, a number of computationally tractable theoretical heat diffusivity models which are widely used for transport modeling have been evaluated using the same experimental data from which the experimental heat diffusivities were inferred. We note the significant ongoing effort to model transport processes with large- scale gyro-kinetic or gyro-fluid computer simulations of turbulent transport (e.g. Ref. 6). Such calculations will in the future be able to provide a rigorous test of turbulent transport mechanisms against experiment. However, such calculations for the plasma edge (including the various atomic physics, radiation and other edge phenomena) are not yet widely available. Thus, we were motivated to undertake a comparison of experimentally inferred heat diffusivities in the edge of DIII-D with the predictions of computationally tractable theoretical models evaluated also using that experimental data, with the intent of obtaining qualitative and semi-quantitative physical insights that can