Infinite Boundary Friction Limit for Weak Solutions of the Stochastic Navier-Stokes Equations

Daniel Goodair · 2025

Read Infinite Boundary Friction Limit for Weak Solutions of the Stochastic Navier-Stokes Equations by Daniel Goodair (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2504-06238.txt.

Introduction

Daniel Goodair studies what happens to a stirred, swirling fluid as friction at its walls grows without bound. He works with the two-dimensional stochastic Navier-Stokes equations, the standard math for fluid motion with noise added, under boundary rules that let the fluid slip as the friction allows.

He proves that as wall friction is pushed to infinity, the moving solutions settle onto one limit. That limit is the single solution of the same system under a strict no-slip rule, where the fluid sticks to the wall. The result holds when the starting flow is known only in a rough, low-detail sense. The author notes the claim seems first of its kind for noisy flow, and new even for quiet flow with such rough starts. The noise in the main result stretches and carries the flow, though other noise types also work.

Catalog record

CREATORDaniel Goodair
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE69 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2504-06238

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Infinite Boundary Friction Limit for Weak Solutions of the Stochastic Navier-Stokes Equations Daniel Goodair Abstract We address convergence of the unique weak solutions of the 2D stochastic Navier-Stokes equations with Navier boundary conditions, as the boundary friction is taken uniformly to infinity, to the unique weak solution under the no-slip condition. Our result is that for initial velocity in L x 2 L^{2}_{x} , the convergence holds in probability in C t ​ W x − ε , 2 ∩ L t 2 ​ L x 2 C_{t}W^{-\varepsilon,2}_{x}\cap L^{2}_{t}L^{2}_{x} for any 0 < ε 0<\varepsilon . The noise is of transport-stretching type, although the theorem holds with other transport, multiplicative and additive noise structures. This seems to be the first work concerning the large boundary friction limit with noise, and convergence for weak solutions, due to only L x 2 L^{2}_{x} initial data, appears new even deterministically. Hello 1 Introduction We are concerned with the 2D stochastic Navier-Stokes equations in a smooth bounded domain \mathscr{O} under Navier boundary conditions, which read as u\cdot\underline{n}=0,\qquad 2(Du)\underline{n}\cdot\mathbf{\underline{\iota}}+\alpha u\cdot\mathbf{\underline{\iota}}=0 (1) where u represents the fluid velocity, \underline{n} is the unit outwards normal vector, \underline{\iota} the unit tangent vector, Du is the rate of strain tensor (Du)^{k,l}:=\frac{1}{2}\left(\partial_{k}u^{l}+\partial_{l}u^{k}\right) and \alpha\in C^{2}(\partial\mathscr{O};\mathbb{R}) represents a friction coefficient which determines the extent to which the fluid slips on the boundary relative to the tangential stress. These conditions were first proposed by Navier in [ 63 , 64 ], observed in [ 58 ] from the kinetic theory of gases and rigorously derived in [ 57 ] as a hydrodynamic limit of the Boltzmann equation under kinetic boundary conditions. Furthermore they have proven viable for modelling rough boundaries, whereby the irregular boundary is smoothened and supplemented with a homogenised boundary condition reflecting the average effect of the rough boundary; this avenue is explored in [ 3 , 31 , 43 ], and may well have taken motivation in earlier works such as [ 66 ] where the Navier boundary conditions are placed on an artificially imposed boundary within the boundary layer to capture the turbulent dynamics there, following the k\sim\varepsilon model of turbulence introduced in [ 49 ]. In this regime, for a smooth domain, \alpha is explicitly defined in terms of the velocity: otherwise, \alpha depends on the roughness of the boundary. Experimental evidence supporting that boundary roughness dominates the dynamics is given in [ 82 ]. Of course conditions ( 1 ) are less classical in the study of the Navier-Stokes equations than the no-slip condition, given simply by u=0 (2) on the boundary \partial\mathscr{O} . Formally we can reconcile these notions by taking the boundary friction \alpha to infinity in ( 1 ), as then u\cdot\underline{\iota} dominates the second equation and the whole of ( 1 ) reduces to ( 2 ). The goal of this paper is to rigorously address the convergence of solutions under Navier boundary conditions ( 1 ) to the solution without slip at the boundary ( 2 ), in the presence of noise. 1.1 The Model Specifically we shall consider the Navier-Stokes equations with a transport-stretching noise, following the principle of Stochastic Advection by Lie Transport introduced in [ 40 ], given by u_{t}=u_{0}-\int_{0}^{t}\mathcal{P}\mathcal{L}_{u_{s}}u_{s}\ ds-\nu\int_{0}^{t}Au_{s}\,ds+\frac{1}{2}\int_{0}^{t}\sum_{i=1}^{\infty}\mathcal{P}B_{i}^{2}u_{s}ds-\int_{0}^{t}\mathcal{P}Bu_{s}d\mathcal{W}_{s}. (3) Here \mathcal{P} denotes the Leray Projector onto divergence-free vector fields tangent to the boundary, A=-\mathcal{P}\Delta is the Stokes Operator and \mathcal{L} represents the nonlinear term defined for sufficiently regular functions f,g:\mathscr{O}\rightarrow\mathbb{R}^{2} by \mathcal{L}_{f}g:=\sum_{j=1}^{2}f^{j}\partial_{j}g , where the superscript notation is for the j^{\textnormal{th}}- component mapping. As for the noise, \mathcal{W} is a Cylindrical Brownian Motion and B is defined with respect to a collection of divergence-free vector-fields (\xi_{i}) by B_{i}f:=\mathcal{L}_{\xi_{i}}f+\mathcal{T}_{\xi_{i}}f:=\sum_{j=1}^{2}\left(\xi_{i}^{j}\partial_{j}f^{j}+f^{j}\nabla\xi_{i}^{j}\right). (4) The noise is introduced as a Stratonovich integral, at the non-projected level of the equation where the pressure appears; the \sum_{i=1}^{\infty}\mathcal{P}B_{i}^{2} term arises as the Itô-Stratonovich corrector after first projecting the equation by \mathcal{P} . We note use of the property that \mathcal{P}B_{i}\mathcal{P}=\mathcal{P}B_{i} hence (\mathcal{P}B_{i})^{2}=\mathcal{P}B_{i}^{2} observed in [ 36 ] Lemma 2.7, and defer to [ 36 ] Proposition 3.2 for the full conversion. Our motivation for considering this noise structure is threefold. Firstly is the physical significance, which is entirely the motivation of [ 40 ]. In [ 40 ] the noise is derived through geometric variational principles and is shown to preserve Kelvin’s Circulation Theorem. This theory has been expanded upon across [ 16 , 41 , 74 ], and has run in tandem with derivations of transport noise in fluids through a Lagrangian Reynolds Decomposition and Transport Theorem given by Mémin [ 59 ] and further developed in [ 8 , 18 , 68 ]. The theory is bolstered by numerical analysis and data assimilation presented throughout [ 7 , 12 , 13 , 14 , 15 , 20 , 21 ] amongst many others. Stratonovich transport noise has also been derived following a stochastic model reduction scheme in [ 19 , 27 , 28 ]. All of this recent progress supports the classical ideas of [ 5 , 48 , 60 , 61 ], and we suggest [ 26 ] for a review of the topic. Secondly is the fact that our choice of noise is very challenging, so proving this difficult case encompasses many others. Well-posedness results which are foundational for the deterministic Euler and Navier-Stokes equations remain open in this stochastic case, particularly in the presence of a boundary: in both the stochastic Euler and Navier-Stokes equations, for their most classical boundary condition, the existence of global strong solutions in 2D and local strong solutions in 3D is unknown. This is in contrast to a multiplicative noise which is (locally) bounded and Lipschitz on the relevant function spaces, where the corresponding theory is given in [ 32 , 33 ]. Interestingly, however, the existence of strong solutions of ( 3 ) under Navier boundary conditions was proven in [ 35 ], so the seemingly more complicated boundary condition acts favourably in the presence of noise. Fortunately the existence of weak solutions is known in both cases, which are the objects of this paper. We only discuss the open questions to motivate the analytical challenges in place with transport-type noise, and as a precursor to say that our results continue to hold with a bounded and Lipschitz multiplicative noise, additive noise in the projected L^{2}_{x} space, and indeed a purely transport noise (without stretching term \mathcal{T}_{\xi_{i}} ). Thirdly we draw attention to the potential regularising properties of transport noise. This was brought to prominence in [ 24 ] where Flandoli, Gubinelli and Priola prove that transport noise restores uniqueness in the transport equation. The theory was given a new lease of life from the work of Galeati [ 29 ], demonstrating that one can choose a transport noise such that the stochastic equation is close to the deterministic equation with added dissipation. In consequence the favourable properties of the deterministic equation with enhanced dissipation can be leveraged, leading in some cases to blow-up control [ 2 , 23 , 25 ] and uniqueness [ 11 , 30 ]. We are keen to highlight this direction due to the relationship that the Navier boundary conditions have with the no-slip condition and the inviscid limit, along with the critical open problems regarding the

Read the complete text