How "the Tail Wags the Dog": Physics of Edge-Core Coupling by Inward Turbulence Propagation

The dynamics of edge-core coupling is critically important to optimal plasma performance. To this end, the physics of what sets the width of edge-core coupling region—where standard Fickian gyrokinetic models sometimes fail—remains an important “known unknown” (also referred to as the “shortfall problem”). Since early proposals by B.B. Kadomtsev, there has been persistent speculation that inward propagation of turbulence from the boundary is a possible means to energize the bulk plasma. Yet, the detailed mechanism of “how the tail wags the dog” has remained a mystery until recent experiments (including BES studies) observed that regular, intense gradient relaxation events (GREs) generated blob-void pairs near the separatrix. Blobs (density excesses) propagated outward into the SOL, while voids (density depletions) propagated inward, thus coupling edge and core plasmas. Here, by developing a model incorporating voids into turbulence dynamics, we demonstrate that this heretofore ignored process of void generation can drive substantial inward turbulence spreading, leading to the formation of a broad turbulent layer.



The physical picture in this model is the Cherenkov emission of drift waves from inward-moving voids. After being generated from GREs, such voids, acting like dressed test particles, will excite a drift wave “radiation field”. In the near field region, the void exhibits an interchange response, which converts to a drift wave in the far field region. The void-excited turbulence could penetrate deeper than the voids themselves and stir the core plasma. This turbulence is regulated by self-generated zonal flow, which may qualitatively account for the observed zonal flow power bursts following the detection of voids in experiments. For typical parameters, the nonlocal inward turbulence spreading is comparable to the local turbulence production, yielding an edge-core coupling region of width ~100 ρ_s. Concomitantly, the (ambient) turbulence and zonal flow can smear or shear the void, thereby constraining the void lifetime τ_v. This process can be characterized by a simple model based on diffusion scattering, which predicts a τ_v ranging from a few to 100 μs, in accord with observation.