Almost anomalous dissipation in advection-diffusion of a divergence-free passive vector

TL;DR

This paper constructs a velocity field in a passive vector advection-diffusion model that exhibits near-anomalous dissipation scaling similar to turbulent flows, bridging mathematical theory and turbulence phenomenology.

Contribution

It introduces a novel construction of velocity fields causing near-anomalous dissipation in passive vector equations, extending the understanding of turbulence-like dissipation behavior.

Findings

Energy dissipation scales as $( ext{log } u^{-1})^{-2}$

Both velocity and passive vector exhibit near-anomalous dissipation

Results align with phenomenological turbulence theories

Abstract

We explore the advection-diffusion of a passive vector described by $\partial_t u + U \cdot \nabla u = - \nabla p + \nu \Delta u$, where both $U$ and $u$ are divergence-free velocity fields. We approach this equation from an input/output perspective, with $U$ as the input and $u$ as the output. This input/output viewpoint has been widely applied in recent studies on passive scalar equation, in the context of anomalous dissipation, mixing, optimal scalar transport, and nonuniqueness problems. What makes the passive vector equation considerably more challenging compared to the passive scalar equation is the lack of a Lagrangian perspective due to the presence of pressure. In this paper, rather than requiring $U$ and $u$ to be identical (as in the Navier-Stokes equation), we require $U$ and $u$ to be identical only in certain characteristics. We focus on the case where the characteristics in question are anomalous and enhanced dissipation. We study the advection-diffusion of a passive vector in a Couette flow configuration. The main result of this paper is a construction of the velocity field $U$ for which the energy dissipation scales as $(\log \nu^{-1})^{-2}$ such that the energy dissipation in velocity field $u$ scales at least as $(\log \nu^{-1})^{-2}$. This means that both $U$ and $u$ exhibit near-anomalous dissipation, where the rate of energy dissipation decreases more slowly than any power-law $\nu^{\alpha}$ for any $\alpha > 0$. The result in this paper is not just a mathematical construct; it closely resembles the behavior of turbulent flow in a channel. The $(\log \nu^{-1})^{-2}$ decrease of energy dissipation is predicted by phenomenological theories of wall-bounded turbulence, a prediction that has been extensively validated through experiments and numerical simulations.