[Paper Review] Fast rotating flows in high spatial dimensions

Jian-Zhou Zhu|arXiv (Cornell University)|May 28, 2019

Fluid Dynamics and Turbulent Flows52 references4 citations

TL;DR

This paper generalizes the Taylor-Proudman theorem (TPT) to compressible flows and $d$-dimensional Euclidean space ($\mathbb{E}^d$, $d \geq 3$) using Lie’s theory and differential forms, establishing a geometric foundation for fast rotating flows. It shows that higher-order corrections to TPT are sub-dominant and do not alter the leading-order behavior, unifying inertial waves, reduced models, and passive scalar dynamics in high dimensions.

ABSTRACT

The central result about fast rotating-flow structures is the Taylor-Proudman theorem (TPT) which connects various aspects of the dynamics. Taylor's geometrical proof of TPT is reproduced and extended substantially, with Lie's theory for general frozen-in laws and the consequent generalized invariant circulation theorems, to compressible flows and to $d$-dimensional Euclidean space ($\mathbb{E}^{d}$) with $d\ge 3$. The TPT relatives, the reduced models (with particular interests on passive-scalar problems), the inertial (resonant) waves and the higher-order corrections, are discussed coherently for a comprehensive bird view of rotating flows in high spatial dimensions.

Motivation & Objective

• To extend the Taylor-Proudman theorem (TPT) to compressible flows and $d$-dimensional Euclidean space ($\mathbb{E}^d$, $d \geq 3$), generalizing its geometric and topological foundations.
• To unify the treatment of reduced models, inertial (resonant) waves, and passive scalar dynamics in high-dimensional rotating flows using a common geometric framework.
• To investigate whether fast rotation in higher dimensions can generate cylinder conditions analogous to those in 3D, enabling passive scalar dynamics in reduced models.
• To clarify the role of higher-order corrections in the circulation theorems and their consistency with leading-order TPT behavior.
• To establish a systematic, geometry-based approach to rotating flows that generalizes Taylor’s original insight beyond incompressible, 3D flows.

Proposed method

• Reproduces and extends Taylor’s geometrical proof of TPT using Lie’s theory for general frozen-in laws and generalized invariant circulation theorems.
• Applies differential forms and Stokes' lemma to derive the circulation invariance in $\mathbb{E}^d$, linking Helmholtz and Kelvin-type theorems in higher dimensions.
• Derives the asymptotic TPT for compressible flows as $\partial_z \bm{u}_h \to \bm{0}$ and $\nabla_h \cdot \bm{u}_h \to 0$, with no constraint on $\partial_z u_z$, under fast rotation.
• Analyzes higher-order corrections via $\verb"d"\verb"U"_R \wedge \verb"d"\verb"U"$ and $L_u \verb"d"\verb"U" \wedge \verb"d"\verb"U"_R$, showing they vanish in the leading-order circulation theorems.
• Uses the time-dependent version of the reduced model and the cylinder condition in $\mathbb{E}^5$ to demonstrate consistency of passive scalar dynamics under multiple rotation symmetries.
• Demonstrates that sub-dominant corrections in $\Omega_R$ correspond to second-order terms in $\bm{u}$, which do not affect the first-order TPT structure.

Experimental results

Research questions

• RQ1 Can the Taylor-Proudman theorem be generalized to compressible flows in $d$-dimensional Euclidean space ($\mathbb{E}^d$, $d \geq 3$) using geometric and topological methods?
• RQ2 Do inertial waves and reduced models in high-dimensional rotating flows arise from the same geometric constraints as in 3D?
• RQ3 Is the passive scalar dynamics in 3D Navier-Stokes flows equivalent to a dimensional reduction from a 5D flow with cylinder conditions under fast rotation?
• RQ4 Are higher-order corrections to the circulation theorems consistent with the leading-order Taylor-Proudman result, or do they introduce new physics?
• RQ5 Can the geometric framework of Lie’s theory and differential forms unify the treatment of frozen-in laws, circulation theorems, and reduced dynamics in high-dimensional rotating systems?

Key findings

• The Taylor-Proudman theorem for compressible flows in $\mathbb{E}^d$ ($d \geq 3$) is generalized via Lie’s theory and differential forms, yielding $\partial_z \bm{u}_h \to \bm{0}$ and $\nabla_h \cdot \bm{u}_h \to 0$ under fast rotation.
• The time-dependent version of the reduced model in $\mathbb{E}^3$ is shown to be consistent with the geometric and topological structure of the circulation theorems.
• Higher-order corrections to the circulation theorems, expressed as $\verb"d"\verb"U"_R \wedge \verb"d"\verb"U"$, vanish in the leading-order limit, confirming their irrelevance to the TPT.
• Sub-dominant contributions from $\Omega_R$ are shown to be quadratic in $\bm{u}$ derivatives and thus irrelevant to the first-order TPT, validating the robustness of the leading-order result.
• The cylinder condition in $\mathbb{E}^5$ under double rotation preserves the same structure as in 3D, confirming the consistency of the geometric approach across dimensions.
• The formalism reveals that passive scalar dynamics in 3D can be viewed as a dimensional reduction from a 5D flow with two cylinder conditions, though no new physics emerges beyond known constraints.