Intermittent singular solutions of the stationary 2D Navier-Stokes equations in sharp Sobolev spaces

Estepan Ashkarian, Ataleshvara Bhargava, Nicholas Gismondi, Matthew Novack · 2025

Read Intermittent singular solutions of the stationary 2D Navier-Stokes equations in sharp Sobolev spaces by Estepan Ashkarian, Ataleshvara Bhargava, Nicholas Gismondi, Matthew Novack (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-2506-00841.txt.

Introduction

Estepan Ashkarian, Ataleshvara Bhargava, Nicholas Gismondi, and Matthew Novack build unusual solutions to the stationary Navier-Stokes equations in two dimensions. Their 2025 paper works on the flat torus and in sharp Sobolev spaces.

The solutions they build are not square integrable, so the usual sense of a solution has to be restated. Their result extends recent work by Lemarié-Rieusset to a sharper setting. The new part of the proof is intermittency, where activity is concentrated in small bursts rather than spread out evenly. For readers, the paper shows how weak a solution can be while still being non-trivial.

Catalog record

CREATOREstepan Ashkarian, Ataleshvara Bhargava, Nicholas Gismondi, Matthew Novack
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE50 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2506-00841

Browse the full library · About the Tower of Babel

Full text

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

Intermittent singular solutions of the stationary 2D Navier-Stokes equations in sharp Sobolev spaces Estepan Ashkarian, Ataleshvara Bhargava, Nicholas Gismondi, Matthew Novack Abstract In this paper we construct non-trivial solutions to the stationary Navier-Stokes equations on the two dimensional torus which lie in ⋂ ϵ ∈ ( 0 , 1 ) L 2 − ϵ ​ ( 𝕋 2 ) ∩ H ˙ − ϵ ​ ( 𝕋 2 ) \bigcap_{\epsilon\in(0,1)}L^{2-\epsilon}(\mathbb{T}^{2})\cap\dot{H}^{-\epsilon}(\mathbb{T}^{2}) . Due to the fact that our solutions are not square integrable, we must redefine the notion of solution. Our result gives a sharp extension of the recent work of Lemarié-Rieusset, [ 19 ] who proved a similar result in the space H ˙ − 1 ∩ BMO − 1 \dot{H}^{-1}\cap\operatorname{BMO}^{-1} . The main new ingredient is the incorporation of intermittency into the construction of the solutions. 1 Introduction 1.1 Motivation and background In this paper we consider the stationary two-dimensional incompressible Navier-Stokes equations \begin{cases}\operatorname{div}(u\otimes u)-\Delta u+\nabla p=0\\ \operatorname{div}(u)=0\end{cases} (1.1) posed on the torus for u(x)\colon\mathbb{T}^{2}\to\mathbb{R}^{2} . We will consider mean zero solutions, that is \int_{\mathbb{T}^{2}}u\,dx=0. Let us start by recalling known results on solutions of ( 1.1 ). If u is a smooth solution of ( 1.1 ), then it must be a trivial one. Indeed, we multiply the equation by u , integrate by parts, and use \operatorname{div}(u)=0 to obtain \displaystyle\sum_{i=1}^{2}\|\nabla u^{i}\|_{L^{2}(\mathbb{T}^{2})}^{2} \displaystyle=\int_{\mathbb{T}^{2}}\partial_{j}u^{i}\partial_{j}u^{i}\,dx \displaystyle=-\int_{\mathbb{T}^{2}}u^{i}\partial_{jj}u^{i}\,dx \displaystyle=-\int_{\mathbb{T}^{2}}\partial_{j}(u^{i}u^{j})u^{i}\,dx-\int_{\mathbb{T}^{2}}u^{i}\partial_{i}p\,dx \displaystyle=0\,, which implies for mean-zero solutions that u\equiv 0 . The same computation can be justified when we assume u is a solution of ( 1.1 ) in L^{p}(\mathbb{T}^{2}) when p>2 . In this scenario, u must in fact be smooth, and hence we can apply the previous argument to conclude u\equiv 0 ; see recent work of Lemarié-Rieusset [ 19 ] for more details. Moreover, in [ 19 ] it is shown that any solution u of ( 1.1 ) in the Lorentz space L^{2,1}(\mathbb{T}^{2}) is trivial. On the other hand, in [ 19 ] Lemarié-Rieusset proved that there exist nontrivial stationary solutions in \dot{H}^{-1}(\mathbb{T}^{2}) to ( 1.1 ). While it is clear how to define weak solutions to ( 1.1 ) if u\in L^{2}(\mathbb{T}^{2}) , it is not obvious what is meant by a weak solution if u\otimes u is not locally integrable. Lemarié-Rieusset treats this issue by expanding u\otimes u in a doubly infinite Fourier series and checking that this infinite sum converges in a negative Sobolev space. We shall adopt a similar notion in Definition 1.2 using paraproducts. While the result in [ 19 ] is the first for nontrivial stationary weak solutions, there are two earlier but relevant results of Christ [ 12 , 13 ] . In [ 12 ] , Christ constructs nontrivial time-dependent solutions of the one dimensional periodic cubic nonlinear Schrödinger equation which belong to C^{0}_{t}H^{-\epsilon}_{x} . The key issue is again the interpretation of the cubic nonlinearity |u|^{2}u when u\notin L^{3}_{t,x} , and Christ shows that this object has an interpretation as a limit \lim_{\epsilon\rightarrow 0}|u_{\epsilon}|^{2}u_{\epsilon} , where u_{\epsilon} is a mollified version of u . In [ 13 ] , Christ then treats the time-dependent two-dimensional Navier-Stokes equation using similar methodology. We remark that Christ identifies a connection between his work and the well-known earlier works of Scheffer [ 21 ] and Shnirelman [ 22 ] on non-uniqueness for the Euler equations; in particular, we quote the following observation from [ 13 ] : Our construction and that of Shnirelman have in common both the use of driving forces tending weakly to zero, and the exploitation of this reverse energy cascade. In particular this also highlights the connection between Christ’s works [ 12 , 13 ] and modern convex integration/Nash iteration techniques descended from [ 21 , 22 ] . 1.2 Main result In this paper, we consider the following notion of stationary solution. For f\in H^{s} and g\in H^{s^{\prime}} , s,s^{\prime}\in\mathbb{R} , we formally define fg=\sum_{j,j^{\prime}\geq 0}\mathbb{P}_{2^{j}}f\mathbb{P}_{2^{j^{\prime}}}g. The projection operators \mathbb{P}_{2^{j}} are the standard Littlewood-Paley projections onto frequencies of size \approx 2^{j} ; see Definition 2.1 . There is not much that can be said in general about the convergence of this sum in negative Sobolev spaces. However, we may offer the following definition for the mean-zero portion \mathbb{P}_{\neq 0}(fg)=(\operatorname{Id}-\fint_{\mathbb{T}^{2}})(fg) of a formal product fg . Definition 1.1 ( Paraproducts in \dot{H}^{s}(\mathbb{T}^{2}) ) . Let f,g be distributions, so that \mathbb{P}_{2^{j}}(f),\mathbb{P}_{2^{j^{\prime}}}(g) are well-defined for j,j^{\prime}\geq 0 . We say that \mathbb{P}_{\not=0}(fg) is well-defined as a paraproduct in \dot{H}^{s}(\mathbb{T}^{2}) for some s\in\mathbb{R} if \sum_{j,j^{\prime}\geq 0}\left\|\mathbb{P}_{\not=0}\left(\mathbb{P}_{2^{j}}(f)\mathbb{P}_{2^{j^{\prime}}}(g)\right)\right\|_{\dot{H}^{s}}<\infty\,. Then we define \mathbb{P}_{\not=0}(fg)=\sum_{j,j^{\prime}\geq 0}\mathbb{P}_{\not=0}\left(\mathbb{P}_{2^{j}}(f)\mathbb{P}_{2^{j^{\prime}}}(g)\right)\,, since the right-hand side is an absolutely summable series in \dot{H}^{s}(\mathbb{T}^{2}) . With the above definition in hand, we define a weak notion of solution to ( 1.1 ). Definition 1.2 ( Weak paraproduct solutions to ( 1.1 ) ) . If u\in\dot{H}^{s} , s<0 , we say u is a weak paraproduct solution to ( 1.1 ) if there is s^{\prime}\in\mathbb{R} such that \mathbb{P}_{\not=0}(u\otimes u) is well defined as a paraproduct in \dot{H}^{s^{\prime}} in the sense of Definition 1.1 and \left\langle\Delta\phi^{i},u^{i}\right\rangle_{\dot{H}^{-s},\dot{H}^{s}}+\left\langle\partial_{j}\phi^{i},\mathbb{P}_{\not=0}(u^{i}u^{j})\right\rangle_{\dot{H}^{-s^{\prime}},\dot{H}^{s^{\prime}}}=0 for all smooth, divergence free vector fields \phi . Our main result is then as follows. Theorem 1.3 ( Non-trivial solutions of 2D stationary Navier-Stokes equations ) . There exist nontrivial (non-constant and non-zero) weak solutions to ( 1.1 ) in the sense of Definition 1.2 belonging to \displaystyle\bigcap_{\epsilon\in(0,1)}L^{2-\epsilon}(\mathbb{T}^{2})\cap\dot{H}^{-\epsilon}(\mathbb{T}^{2}) . Considering that stationary mean-zero solutions in L^{2,1}(\mathbb{T}^{2}) must be trivial from [ 19 ] , Theorem 1.3 is essentially sharp. The only possible strengthening would be to address the endpoint case u\in L^{2}(\mathbb{T}^{2}) . It appears that there are clear obstructions which prevent any simple extension of our method to the case u\in L^{2}(\mathbb{T}^{2}) . However, it is conceivable to us that stationary solutions in L^{2}(\mathbb{T}^{2}) may be nontrivial, although a different method of proof may be required. Remark 1.4 ( Extensions of Theorem 1.3 ) . The same techniques we use to prove Theorem 1.3 can easily be extended to prove the following: let u_{0} be any smooth divergence-free vector field. Then for any \epsilon>0 , there exists u_{\epsilon},p_{\epsilon} which solves ( 1.1 ) in the sense of Definition 1.2 and satisfies \|u_{\epsilon}-u_{0}\|_{\dot{H}^{-\epsilon}}+\|u_{\epsilon}-u_{0}\|_{L^{2-\epsilon}}<\epsilon . Remark 1.5 ( Regularity in L^{\infty} -based Besov spaces ) . Our construction gives solutions lying in \bigcap_{\epsilon>0}B^{-\nicefrac{{1}}{{2}}-\epsilon}_{\infty,\infty} . Indeed this is immediate from Bernstein’s inequality and the absolute summability of the series used to define u=\sum_{q\geq 0}u_{q+1}-u_{q} in \bigcap_{\epsilon>0}L^{2-\epsilon} ; see ( 3.1 ). 1.3 Related literature In addition to the recent work of Lemarié-Rieusset [ 19 ] and the earlier work of Christ [ 13 ] , the past decade has seen numerous flexibility results for the Navier-Stokes equations. Many of these fall under the umbrella of convex integration/Nash iteration, initiated by De Lellis and Székelyhidi in [ 16 , 17 ] for the Euler equations. Most relevant to our setting is the intermittent style of Nash iteration, developed first by Buckmaster and Vicol [ 7 ] for the Navier-Stokes equations. Variants of intermittent convex integration have been utilized by Luo [ 20 ] for the 4D NSE, Buckmaster, Colombo, and Vicol [ 6 ] to concentrate the temporal sets on which the solutions to 3D NSE are rough, and Cheskidov and Luo [ 10 , 11 ] for solutions in L^{2-}_{t}L^{\infty}_{x} and

Read the complete text