Configuration of the GPI viewing arrays; (a) side view and (b) top view. The solid (red) D-shape curve represents the LCFS. Also marked are the toroidal and poloidal angles of separation between the two viewing arrays.
Density fluctuations in the low-field-side edge of an I-mode plasma at r/a = 0.95; conditional wavenumber and frequency spectrum . The WCM appears as the broad feature at and , with a lab frame propagation in the electron diamagnetic drift direction.
Poloidally averaged Fourier spectra of TDE poloidal velocities in an I-mode (solid blue) and the preceding L-mode (dashed grey). The sample size for correlation measurements is , the spectra are time averaged from a spectrogram with df = 0.5 kHz.
Cross-coherency and cross-phase (a) between the top and bottom TDE view-pairs on the outboard side GPI array and (b) between the inboard and outboard GPI arrays. The latter are conditionally averaged for the times at which the fluctuation level is sufficiently large for the TDE routine to reliably lock into propagating features.
GAM frequencies against the theoretical from a large number of experiments, with the recent results from Alcator C-Mod overlaid as the red squares. The dashed line represents . Original reprinted with permission from A. Fujisawa et al., Nucl. Fusion 47, S718 (2007). Copyright (2007) International Atomic Energy Agency.
Radial location of the edge turbulence features in I-mode; (top) radial profile of the mode amplitude of the WCM (squares, right axis) and the GAM (full circles, left axis) exhibiting nearly identical distributions; (bottom) radial electric field profile in the same I-mode discharge as above in a sample including the time of the velocity measurement. Profile is calculated from CXRS measurements.
Time history of a long lived I-mode discharge; (a) core electron temperature, (b) electron temperature at the GAM measurement location, i.e., the top of the pedestal, (c) spectrogram of the TDE poloidal velocity , (d) spectrogram of density fluctuations in the same spatial location restricted to a wavenumber range to highlight the WCM. (a) and (b) are measured via ECE and (c) and (d) by GPI.
Nonlinear spectral transfer of density fluctuation power; (a) nonlinear transfer as a function of source frequency f 1 and target frequency f, with a overlaid as the white dashed line, and the boundaries of meaningful velocity measurements marked by dashed grey lines; (b) transfer function , with a clear positive peak at ; (c) nonlinear transfer normalized to the spectral power of source and target components, with the horizontal dashed line marking the central frequency of the WCM; (d) effective transfer rate, i.e., transfer function normalized to the spectral power with a clear negative peak at the WCM frequency marked in plot (c). High frequency noise is clipped at f = 250 kHz in part (d).
Nonlinear kinetic energy transfer function (left) and the effective nonlinear growth rate .
Time history of a typical L-I transition followed by an I-H transition event; graphs are from top to bottom: the solid black line represents the estimated, time-resolved, non-linear GAM drive , the red curve corresponds to estimated from ECE and Thompson scattering, the middle panel shows , and the bottom .
Article metrics loading...
Full text loading...