Prof Michael Barnes

Abstract:

The theory of magnetohydrodynamic (MHD) turbulence predicts that Alfvénic and slow-mode-like compressive fluctuations are energetically decoupled at small scales in the inertial range. The partition of energy between these fluctuations determines the nature of dissipation, which, in many astrophysical systems, happens on scales where plasma is collisionless. However, when the magnetorotational instability (MRI) drives the turbulence, it is difficult to resolve numerically the scale at which both types of fluctuations start to be decoupled because the MRI energy injection occurs in a broad range of wavenumbers, and both types of fluctuations are usually expected to be coupled even at relatively small scales. In this study, we focus on collisional MRI turbulence threaded by a near-azimuthal mean magnetic field, which is naturally produced by the differential rotation of a disc. We show that, in such a case, the decoupling scales are reachable using a reduced MHD model that includes differential-rotation effects. In our reduced MHD model, the Alfvénic and compressive fluctuations are coupled only through the linear terms that are proportional to the angular velocity of the accretion disc. We numerically solve for the turbulence in this model and show that the Alfvénic and compressive fluctuations are decoupled at the small scales of our simulations as the nonlinear energy transfer dominates the linear coupling below the MRI-injection scale. In the decoupling scales, the energy flux of compressive fluctuations contained in the small scales is almost double that of Alfvénic fluctuations. Finally, we discuss the application of this result to prescriptions of ion-to-electron heating ratio in hot accretion flows.

Abstract:

Abstract Ion-gyroradius-scale microinstabilities typically have a frequency comparable to the ion transit frequency. Due to the small electron-to-ion mass ratio and the large electron transit frequency, it is conventionally assumed that passing electrons respond adiabatically in ion-gyroradius-scale modes. However, in gyrokinetic simulations of ion-gyroradius-scale modes in axisymmetric toroidal magnetic fields, the nonadiabatic response of passing electrons can drive the mode, and generate fluctuations in narrow radial layers, which may have consequences for turbulent transport in a variety of circumstances. In flux tube simulations, in the ballooning representation, these instabilities reveal themselves as modes with extended tails. The small electron-to-ion mass ratio limit of linear gyrokinetics for electrostatic instabilities is presented, in axisymmetric toroidal magnetic geometry, including the nonadiabatic response of passing electrons and associated narrow radial layers. This theory reveals the existence of ion-gyroradius-scale modes driven solely by the nonadiabatic passing electron response, and recovers the usual ion-gyroradius-scale modes driven by the response of ions and trapped electrons, where the nonadiabatic response of passing electrons is small. The collisionless and collisional limits of the theory are considered, demonstrating parallels in structure and physical processes to neoclassical transport theory. By examining initial-value simulations of the fastest-growing eigenmodes, the predictions for mass-ratio scaling are tested and verified numerically for a range of collision frequencies. Insight from the small electron-to-ion mass ratio theory may lead to a computationally efficient treatment of extended modes.

Abstract:

A linear response, local model for the DBS amplitude applied to gyrokinetic simulations shows that radial correlation Doppler reflectometry measurements (RCDR, Schirmer et al 2007 Plasma Phys. Control. Fusion 49 1019) are not sensitive to the average turbulence radial correlation length, but to a correlation length that depends on the binormal wavenumber k⊥ selected by the Doppler backscattering (DBS) signal. Nonlinear gyrokinetic simulations show that the turbulence naturally exhibits a nonseparable power law spectrum in wavenumber space, leading to a power law dependence of the radial correlation length with binormal wavenumber lr∼Ck⊥-α(α≈1) which agrees with the inverse proportionality relationship between the measured lr and k⊥ observed in experiments (Fernández-Marina et al 2014 Nucl. Fusion 54 072001). This new insight indicates that RCDR characterizes the eddy aspect ratio in the perpendicular plane to the magnetic field. It also motivates future use of a nonseparable turbulent spectrum to quantitatively interpret RCDR and potentially other turbulence diagnostics. The radial correlation length is only measurable when the radial resolution at the cutoff location Wn satisfies Wn≪lr, while the measurement becomes dominated by Wn for Wn≫lr . This suggests that lr is likely to be inaccessible for electron-scale DBS measurements (k⊥ρs>1 ). The effect of Wn on ion-scale radial correlation lengths could be nonnegligible.

Abstract:

A linear response, local model for the DBS amplitude applied to gyrokinetic simulations shows that radial correlation Doppler reflectometry measurements (RCDR, Schirmer et al 2007 Plasma Phys. Control. Fusion 49 1019) are not sensitive to the average turbulence radial correlation length, but to a correlation length that depends on the binormal wavenumber k⊥ selected by the Doppler backscattering (DBS) signal. Nonlinear gyrokinetic simulations show that the turbulence naturally exhibits a nonseparable power law spectrum in wavenumber space, leading to a power law dependence of the radial correlation length with binormal wavenumber lr ∼ Ck−α ⊥ (α ≈ 1) which agrees with the inverse proportionality relationship between the measured lr and k⊥ observed in experiments (Fern´andez-Marina et al 2014 Nucl. Fusion 54 072001). This new insight indicates that RCDR characterizes the eddy aspect ratio in the perpendicular plane to the magnetic field. It also motivates future use of a nonseparable turbulent spectrum to quantitatively interpret RCDR and potentially other turbulence diagnostics. The radial correlation length is only measurable when the radial resolution at the cutoff location Wn satisfies Wn ≪ lr , while the measurement becomes dominated by Wn for Wn ≫ lr . This suggests that lr is likely to be inaccessible for electron-scale DBS measurements (k⊥ρs > 1). The effect of Wn on ion-scale radial correlation lengths could be nonnegligible.

Abstract:

Ion-gyroradius-scale microinstabilities typically have a frequency comparable to the ion transit frequency. Due to the small electron-to-ion mass ratio and the large electron transit frequency, it is conventionally assumed that passing electrons respond adiabatically in ion-gyroradius-scale modes. However, in gyrokinetic simulations of ion-gyroradius-scale modes in axisymmetric toroidal magnetic fields, the nonadiabatic response of passing electrons can drive the mode, and generate fluctuations in narrow radial layers, which may have consequences for turbulent transport in a variety of circumstances. In flux tube simulations, in the ballooning representation, these instabilities reveal themselves as modes with extended tails. The small electron-to-ion mass ratio limit of linear gyrokinetics for electrostatic instabilities is presented, in axisymmetric toroidal magnetic geometry, including the nonadiabatic response of passing electrons and associated narrow radial layers. This theory reveals the existence of ion-gyroradius-scale modes driven solely by the nonadiabatic passing electron response, and recovers the usual ion-gyroradius-scale modes driven by the response of ions and trapped electrons, where the nonadiabatic response of passing electrons is small. The collisionless and collisional limits of the theory are considered, demonstrating parallels in structure and physical processes to neoclassical transport theory. By examining initial-value simulations of the fastest-growing eigenmodes, the predictions for mass-ratio scaling are tested and verified numerically for a range of collision frequencies. Insight from the small electron-to-ion mass ratio theory may lead to a computationally efficient treatment of extended modes.