On the relevance of uncorrelated Lorentzian pulses for the interpretation of turbulence in the edge of magnetically confined toroidal plasmas

TL;DR

This study tests the hypothesis that edge turbulence in fusion plasmas can be described by uncorrelated Lorentzian pulses, but finds that actual data shows correlated, power-law distributed pulses, challenging previous assumptions about turbulence spectra.

Contribution

Introduces a novel wavelet-based method for pulse detection and applies it to fusion plasma data, revealing correlated, power-law pulse distributions.

Findings

Detected pulses do not have narrow duration distributions.

Pulses exhibit long-range temporal correlations.

Power-law distributions characterize pulse durations.

Abstract

Recently, it has been proposed that the turbulent fluctuations measured in a linear plasma device could be described as a superposition of uncorrelated Lorentzian pulses with a narrow distribution of durations, which would provide an explanation for the reported quasi-exponential power spectra. Here, we study the applicability of this proposal to edge fluctuations in toroidal magnetic confinement fusion plasmas. For the purpose of this analysis, we introduce a novel wavelet-based pulse detection technique that offers important advantages over existing techniques. It allows extracting the properties of individual pulses from the experimental time series, and quantifying the distribution of pulse duration and energy, as well as temporal correlations. We apply the wavelet technique to edge turbulent fluctuation data from the W7-AS stellarator and the JET tokamak, and find that the pulses detected in the data do not have a narrow distribution of durations and are not uncorrelated. Instead, the distributions are of the power law type, exhibiting temporal correlations over scales much longer than the typical pulse duration. These results suggest that turbulence in open and closed field line systems may be distinct and cast doubt on the proposed ubiquity of exponential power spectra in this context.