April 23-25

Log in

Abstract Details

files Add files


Author: Adrian E Fraser
Requested Type: Pre-Selected Invited
Submitted: 2018-02-28 12:21:10

Co-authors: M.J.Pueschel, P.W.Terry, E.G.Zweibel

Contact Info:
University of Wisconsin-Madison
1150 University Ave
Madison, WI   53706

Abstract Text:
In fusion and astrophysical systems, sheared flows can drive the Kelvin-Helmholtz (KH) instability, leading to turbulence that enhances mixing and momentum transport. In KH-unstable flows, growing modes coexist with damped modes at the same wavenumber. Recent analytical calculations show these large-scale damped modes, traditionally ignored in theories of KH turbulence, play a key role in instability saturation [Fraser et al. PoP (2017)]. Because they are excited nonlinearly, they remove energy from large-scale fluctuations before it cascades to small scales, and can significantly modify turbulent momentum transport. This modifies the usual picture of turbulence as a balance between forcing at large, dissipationless scales and energy transfer to small, dissipative scales, and suggests a route for improving reduced transport models by including stable mode effects.
We present results from nonlinear gyrokinetic simulations of KH-driven turbulence relevant to fusion devices and accretion disks. The flow is maintained by a Krook-like forcing, allowing driven rather than decaying turbulence, and we add a radiative damping term that preferentially damps large-scale fluctuations. When this damping is small, stable modes are not only relevant at saturation onset, but remain so during the quasi-stationary state. As a result, stable modes become an important ingredient for approximations of the turbulent flow and Reynolds stress. Motivated by this observation we use eigenmode decompositions to construct a simple fluid model to describe how mean flow amplitude scales with Krook drive. When stable modes are included, the model exhibits good quantitative agreement with observed scalings. When stable modes are strongly driven but neglected from the model, it fails to describe the system even qualitatively.
Supported by NSF award PHY-1707236, the Wisconsin Alumni Research Foundation, the Vilas Trust, and U.S. DOE award DE-FG02-89ER53291.