Predictive-iterative modelling has been performed to investigate the role of convective and diffusive particle transport in the edge during the density build-up after the L–H transition. For the time-dependent modelling, the 1.5D radial transport code ASTRA has been used. The convective velocity, diffusion coefficient and the particle source profiles have been parameterized. Their parameters were varied until the best match of the modelling to the density measurements was found. The extensive parameter scans show that the density build-up can be reproduced by assuming only a diffusive edge transport barrier (ETB) with reduced diffusion coefficient at the edge with respect to the core values. Moreover, the replacement of the diffusive ETB by a strong inwards directed convective velocity at the edge (edge pinch) did not succeed in describing the data. This indicates that a diffusive ETB is required to explain the density build-up. However, the addition of an edge pinch to the diffusive ETB barrier slightly enhances the agreement between modelling and experiment. The best agreement was found with an edge diffusion coefficient of 0.031 m2 s−1 and an edge convective velocity of −0.5 m s−1. Because of the large uncertainties in the source, it is not possible to pin down the exact value for the additional edge pinch. An upper limit for a possible edge convective velocity of −5 m s−1 was estimated. These findings could also be confirmed by analysing H-mode phases of a collisionality scan, in which the normalized collisionality varied from 3.5 to 5.5 at the pedestal top.
Willensdorfer, M., Fable, E., Wolfrum, E., Aho-Mantila, L., Aumayr, F., Fischer, R., Reimold, F., & Ryter, F. (2013). Particle transport analysis of the density build-up after the L-H transition in ASDEX Upgrade. Nuclear Fusion, 53(9), . https://doi.org/10.1088/0029-5515/53/9/093020