ASF

Configuration-Space Curvature and the Navier–Stokes Singular Set

June 29, 2026|
A
A

Two programmes in mathematics have developed alongside one another for sixty years without meeting. The first, initiated by Arnold in 1966, treats an ideal incompressible fluid as a geodesic on the infinite-dimensional Riemannian manifold of volume-preserving diffeomorphisms and reads the instability of the flow from the curvature of that manifold. The second, developed by Scheffer and by Caffarelli, Kohn and Nirenberg through the 1970s and 1980s, bounds the singular set of a viscous fluid by asking when the energy of the flow concentrates in small parabolic regions. The preprint described in this article shows that the two programmes meet through a single operator identity: the curvature operator of the configuration group is built from the pressure Hessian, the nonlocal transform that drives the dynamics of the velocity gradient. Read against the partial-regularity theory, this identity places the Navier–Stokes singular set where the curvature concentrates or the pressure depletes. At a singular point of Type I the second alternative is excluded.

1. Configuration group of an ideal fluid

The configuration space of an ideal incompressible fluid on a closed Riemannian manifold is the group

of smooth diffeomorphisms that preserve the volume form . A diffeomorphism preserves volume exactly when the fluid is incompressible: no region is compressed or expanded. A configuration of the fluid is a placement of every fluid particle. A path in is therefore a motion of the fluid.

The group has a natural Riemannian metric. At the identity its tangent space is the Lie algebra of smooth divergence-free vector fields on , and the natural inner product there is the pairing

Extending this by right-invariance produces the right-invariant metric on . Right-invariance states that the kinetic energy of the fluid is invariant under relabelling the particles. This is the fundamental symmetry of an ideal fluid. Arnold's observation[1] is that the geodesics of this metric are exactly the solutions of the Euler equations of ideal fluid mechanics.

Theorem (Arnold (1966)).

A smooth path is a geodesic of the right-invariant metric if and only if the Eulerian velocity field satisfies the incompressible Euler equations

The pressure is a Lagrange multiplier for the incompressibility constraint, as in classical mechanics. It is the normal reaction force that keeps the motion on the submanifold inside the ambient group of all diffeomorphisms. For flat the ambient group is flat.

For the incompressible equations on the flat torus we work at the Lie algebra level. The Levi-Civita connection of the metric is the ordinary directional derivative followed by the Leray projection onto the divergence-free part:

The complementary projection onto gradients is what the incompressibility constraint removes from each infinitesimal step. The pressure is what produces.

2. Sectional curvature and the role of pressure

The curvature tensor of the Levi-Civita connection is

Arnold[1] computed the sectional curvature for smooth divergence-free pairs on the two-torus and found it negative in many planes. For the plane spanned by the stream functions and on a two-torus of area , with , his formula gives

where . The formula is a special case of Arnold's formula for planes that contain a shear flow, as restated by Khesin, Langøen and Markina.[16] The curvature is negative unless and are parallel. It does not vanish on a shell : for and it equals . Section 4 derives the three-dimensional form of this formula from the pressure. The sign comes from the pressure: in a plane containing a single Fourier mode the curvature is the negative energy of a pressure gradient built from the pair.

By the Jacobi equation negative sectional curvature drives neighbouring geodesics apart: two Euler flows starting close together in configuration space tend to separate exponentially fast. This is the Lagrangian instability of the ideal flow. Rouchon (1992)[3] and Misiołek (1993)[4] studied this instability through the Jacobi equation and the curvature of the group. Preston (2004)[5] proved that Lagrangian and Eulerian instabilities of ideal fluids are equivalent.

The curvature measures how strongly the incompressibility constraint couples nearby fluid motions. Section 4 expresses it through pressure gradients.

3. Pressure Hessian in viscous dynamics

For the viscous problem on the velocity gradient evolves by

obtained by differentiating the Navier–Stokes momentum equation and using incompressibility to identify the pressure term. The three contributions are the self-stretching , the pressure Hessian , and the viscous term.

The self-stretching is local: it depends only on the velocity gradient at the same point. If the pressure Hessian is replaced by its isotropic part , the equation becomes the Vieillefosse–Cantwell restricted Euler system,[6][7] which blows up in finite time from generic smooth initial data. The anisotropic part of the pressure Hessian is nonlocal and is absent from that system.

The pressure is determined from the flow by the Poisson equation

where is the strain tensor and is the vorticity. The quantity is the strain–enstrophy imbalance. Inverting the Laplacian and differentiating twice gives

where is the matrix Riesz transform with Fourier symbol . The pressure Hessian is a nonlocal zeroth-order Fourier multiplier applied to : its value at a point depends on the distribution of over the entire torus.

This nonlocality makes local models of the velocity-gradient dynamics inadequate. A local closure can at best model the pressure Hessian and discards the information in the configuration-space curvature.

4. Main identity

The curvature of relates to the pressure Hessian through the Gauss equation for the embedding

The ambient group of all smooth diffeomorphisms has a flat metric: its Riemannian curvature vanishes.[2] The curvature of the submanifold is therefore determined entirely by the second fundamental form of the embedding.

For the embedding of inside the second fundamental form is the component of the ambient acceleration orthogonal to (the gradient part produced by the incompressibility projection):

The source is symmetric in . Hence is symmetric. For the source is . Then and . The second fundamental form of the incompressibility constraint is the pressure gradient.

The Gauss equation then gives the curvature of from the shape operator.

Theorem (Main identity).

Let be smooth divergence-free fields on , and let be the pressure of . Let be the mixed pressure determined by . The curvature operator of satisfies

The Gauss equation reproduces Arnold's sectional curvature formula from this operator identity on integration against .

Proof. For a submanifold of a flat space the Gauss equation reads . Take and let range over . For a gradient and divergence-free , one integration by parts gives . With and this yields for every . Leray projection turns this into the stated identity. ∎

The first term is the Leray-projected pressure Hessian applied to . The same matrix field drives the velocity-gradient evolution equation. The configuration-space curvature and the vortex-stretching dynamics share this field.

The second term involves the mixed pressure , whose gradient is . Pairing the identity with , the first term gives and the second gives . Hence . The operator identity says more than the sectional curvature: it identifies the curvature as a specific operator on .

Planes containing a pressureless flow. If then is constant on the torus and . For every the sectional curvature is then

the negative energy of the mixed pressure gradient, with equality exactly when is divergence-free. A single Fourier mode with is pressureless, since . For with and , the projection computed mode by mode on of volume gives

For and in a coordinate plane this reduces to Arnold's formula of Section 2 with in place of . The curvature vanishes exactly when or . This condition does not involve . A spectral computation of both sides of the Gauss equation on a grid agrees with this formula to machine precision. The test set contains the same-shell pair , , where the curvature is negative.

Symbolic and spectral computation in the verify/ directory of the companion repository checks the identities used.

5. Strain–enstrophy imbalance and the pressure Hessian

The main identity determines the curvature operator. Now we measure the pressure Hessian in its first term.

From the Poisson equation , the pointwise trace of the pressure Hessian is

Thus the trace of the pressure Hessian is minus the strain–enstrophy imbalance . The pressure-Hessian term of the curvature vanishes identically for pressureless flows , and Section 4 showed that the curvature in every plane through such a flow is non-positive.

The global size follows from the Riesz isometry.

Theorem (Size of the pressure Hessian).

For a smooth divergence-free on ,

The proof is immediate: because Riesz transforms are isometric on . Thus the norm of the nonlocal pressure Hessian equals the norm of the pointwise imbalance between strain energy and enstrophy. The curvature term is a multiplication operator followed by a projection. Its operator norm on is bounded by . The identity measures the field and gives no bound on that operator norm.

The total imbalance over the torus vanishes at every time,

because the identity holds for every divergence-free field on by integration by parts and incompressibility. On a curved closed manifold the same computation leaves the term . The dynamics redistribute the imbalance and it has no global source.

6. Caffarelli–Kohn–Nirenberg partial-regularity theory

Now the regularity theory of the viscous problem.

A Leray–Hopf weak solution of the Navier–Stokes equations on is a divergence-free field satisfying the equations in the sense of distributions and the global energy inequality. Leray (1934)[8] and Hopf (1951)[9] showed such solutions exist for all time from any initial datum. Whether every smooth initial datum produces a smooth solution for all time is the Clay Millennium Problem.

Where global smoothness is open the partial-regularity theory bounds the set of potential singularities. A point in spacetime is regular if the solution is locally bounded near it; the singular set is the complement. Scheffer (1976)[10] began this theory, and Caffarelli, Kohn and Nirenberg (1982)[11] proved that the one-dimensional parabolic Hausdorff measure of is zero. So has parabolic Hausdorff dimension at most one and contains no rectifiable curve lying in a single time slice.

The central tool is the following criterion. Caffarelli, Kohn and Nirenberg proved it for cylinders shifted slightly forward in time; Ladyzhenskaya and Seregin[17] proved it in the form below, with backwards cylinders.

Theorem (ε-regularity (Caffarelli–Kohn–Nirenberg 1982; Ladyzhenskaya–Seregin 1999)).

There exists an absolute constant such that if is a suitable weak solution and

then is a regular point. Here is the backwards parabolic cylinder. Equivalently, every singular point satisfies the concentration bound

A suitable weak solution satisfies a local energy inequality. This mild additional condition on the Leray–Hopf class is satisfied by solutions constructed by appropriate mollification. The -regularity criterion says that a point is regular if the scaled dissipation (the integral of over a backwards parabolic cylinder divided by its radius) is small at all sufficiently small scales. At a regular point the scaled dissipation tends to zero. The criterion therefore characterises regular points.

A second criterion of the same kind uses the velocity and the pressure. Ladyzhenskaya and Seregin[17] (the first local regularity criterion of their Section 2, rescaled to unit size) show that there is an absolute such that is regular whenever

for one radius and one function . Subtracting from the pressure leaves the equations and the local energy inequality unchanged. Both quantities are invariant under the Navier–Stokes scaling.

7. Curvature concentration and the singular set

The pointwise identity

(since and ) inserts the curvature into the -regularity criterion. Whenever the curvature is small the full gradient energy is controlled by the enstrophy alone.

Theorem (Balance at a singular point).

Let be a suitable weak solution on and let be a singular point. Then either the curvature concentrates,

or the flow comes into strain–enstrophy balance: the scaled strain energy and half-enstrophy share a common positive upper limit,

The proof is a direct consequence of the -regularity criterion. Suppose the curvature does not concentrate. Then . The gradient split gives . Since the second term vanishes . Since the scaled strain energy and half-enstrophy differ by .

The companion coercive statement runs the other way: if the flow remains imbalanced near the singular point (meaning for all small and some fixed ) then non-concentrating curvature forces regularity, because the curvature controls the full gradient energy up to the factor .

Away from balance the curvature governs regularity. At a singular point where the curvature does not concentrate the flow must enter strain–enstrophy balance. Equivalently the pressure depletes locally with in scaled average.

Transport law. The imbalance is transported with sources. Differentiating the Navier–Stokes equation to get the evolution of and tracing gives

The three source terms are the inertial production (the determinant of the velocity gradient, a null Lagrangian that integrates to zero over ), the curvature work (which also integrates to zero by incompressibility and the divergence theorem) and the viscous redistribution. Each term has zero net integral and no term is a global source. This agrees with .

8. Exclusion of balanced singularities of Type I

One regime escapes the curvature: the balanced regime where in scaled average and the pressure depletes. At a singular point of Type I this regime cannot occur.

Type-I hypotheses. Let be a singular point. Write

for the scale-invariant energy, dissipation and pressure. We say is of Type I if for some , and

Condition (T1) is the most common definition of a Type-I singularity. Seregin and Šverák[12] sketch how (T1) bounds scale-invariant energy quantities through the local energy inequality. Here (T1) and (T2) are both assumed. Whether Type-I singular points exist is open: Seregin and Šverák reduce their exclusion to a Liouville conjecture for bounded ancient solutions that remains unproved in three dimensions.

Blow-up limits. The parabolic rescaling at is

This preserves the Navier–Stokes equations, and every quantity above is invariant: , and .

Theorem (Balanced blow-up limit).

Let be a suitable weak solution on with a singular point of Type I, and suppose . Then along a sequence the rescalings converge strongly in for every to a divergence-free field on such that

  1. is smooth, , and solves the pressureless system ;
  2. in the sense of distributions;
  3. does not vanish identically on .

Proof. By (T2) and scaling, is bounded in on each and is bounded in by , uniformly for small . The standard compactness argument of the blow-up method[12][13] gives a sequence with strongly in and weakly in for every . The pair solves the Navier–Stokes equations in the sense of distributions, by weak lower semicontinuity, and (T1) passes to the limit as .

For divergence-free fields . Strong convergence makes converge in . Hence in distributions. The balance hypothesis gives . Hence , and the divergence of the limit equation gives .

For almost every the function is harmonic. The interior gradient estimate for harmonic , integrated over with , gives

So . The scaled pressure bound excludes the term that a harmonic pressure could otherwise have. Then solves the pressureless system in the sense of distributions. Each component satisfies a linear parabolic equation with a drift that is bounded on for every , and parabolic regularity makes smooth.

Suppose on . On split , where and is harmonic in . Riesz transforms are bounded on . Hence , and with independent of . With the gradient estimate gives for . Fix with small, then take large. The quantity in the second -regularity criterion of Section 6 then falls below on for . So is regular for , and is regular for . This contradicts the choice of . ∎

Type-I exclusion. The pressureless system is a uniformly parabolic equation with divergence-free drift and no zeroth-order term. The maximum principle applies to each component.

Lemma (Liouville for the Type-I pressureless limit).

A smooth divergence-free solution of the pressureless system on with is identically zero.

Proof. On each slab with the field is bounded, and the maximum principle for bounded solutions gives . Letting gives . The same argument for gives . ∎

Theorem (No balanced Type-I singularity).

Let be a suitable weak solution on and let be a singular point of Type I. Then is curvature-concentrating,

Proof. Otherwise . The blow-up limit of [thm-blowup] is then a smooth pressureless flow with Type-I decay that does not vanish on , and the lemma forces it to vanish. ∎

Scale-critical class. Replace (T1) by the local bound

and the maximum principle by the Liouville theorem of Seregin, Silvestre, Šverák and Zlatoš[14] for divergence-free drifts in . The resulting statement has no content. The supremum shrinks with . The limit superior is therefore an infimum and the bound says that on some cylinder . The local form of the Escauriaza–Seregin–Šverák theorem[13] then makes regular; Seregin and Šverák[12] list this quantity among those whose boundedness alone implies regularity. No singular point satisfies the local bound. The Type-I bound (T1) does not imply it either: it controls only by . That bound is infinite in the limit .

9. Beyond Type I

Section 8 closes the balanced alternative at singular points of Type I. Without (T1) the rescalings at need not converge to a bounded flow, and the decay that drives the Liouville lemma is lost. If a bounded pressureless limit does arise, it keeps the structure below.

The pressureless structure forces rigid algebraic constraints on the velocity gradient . Incompressibility gives , and the balance eliminates the linear coefficient of the characteristic polynomial. The Cayley–Hamilton theorem then gives

The velocity gradient is either nilpotent (where and there is no inertial production) or its eigenvalues are the three cube roots of . This pattern is determined entirely by a single scalar.

The maximum principle supplies unconditional structure for bounded smooth : for each component , the map is non-increasing and is non-decreasing. The spatial oscillation

is non-increasing in time. A spatially constant pressure gradient shifts the supremum and the infimum by the same amount. The oscillation stays monotone even then. The limit is constant in space at every time if and only if for each component.

The averaged-equation barrier of Tao (2016)[15] shows that the energy identity alone cannot settle such questions. Tao constructed a bilinear operator that preserves the energy cancellation , obeys the same scaling and harmonic-analysis estimates as the Euler operator, and yet admits smooth finite-time blowup. Any argument that uses only the energy identity, scaling and those estimates applies equally to and cannot decide regularity for the true Euler operator . The Riesz identity , the Cayley–Hamilton constraint and the energy supersolution property

use the pointwise form of the Euler nonlinearity. The barrier does not address arguments built on that form.

For the full Navier–Stokes equations, Koch, Nadirashvili, Seregin and Šverák[18] proved that bounded ancient mild solutions are constant in two dimensions and for axisymmetric flows in three dimensions with the decay . The three-dimensional question is open. The pressureless case is a special case of it, and we state it as a conjecture.

Conjecture (Pressureless Liouville).

A bounded smooth divergence-free solution of the pressureless system on is constant.

The conjecture would remove the decay hypothesis from the Liouville lemma. It would not by itself exclude balanced singularities without Type-I bounds. Two gaps remain: bounded blow-up limits at must be produced without (T1), and a nonzero constant limit is invisible to the -regularity argument of Section 8. That argument detects only the size of . A drift that is bounded without decay lies outside the class of Seregin, Silvestre, Šverák and Zlatoš, and their counterexample shows that their condition on the drift is sharp.[14]

10. Summary

The Gauss equation for the embedding gives the curvature of the configuration group from the shape of incompressibility. The ambient group is flat. The whole curvature comes from the second fundamental form of the constraint. This form is the pressure gradient. The curvature operator is the Leray-projected pressure Hessian applied to plus a mixed pressure term. The norm of the pressure Hessian is the norm of the strain–enstrophy imbalance . In every plane through a pressureless flow the sectional curvature is non-positive, and Arnold's single-mode formula is the case of a shear flow.

In the Caffarelli–Kohn–Nirenberg theory the curvature concentrates at a singular point or the flow enters strain–enstrophy balance. Away from balance the curvature controls the local energy coercively and non-concentrating curvature forces regularity. At a singular point of Type I the balanced alternative produces a smooth pressureless ancient flow with Type-I decay, and the maximum principle forces it to vanish: every Type-I singular point is curvature-concentrating. The analogous statement for the local class is empty: the local Escauriaza–Seregin–Šverák theorem makes every such point regular.

Without Type-I bounds the balanced regime is open. The energy identity is blind there, as the Tao barrier shows. The Cayley–Hamilton rigidity, the monotone oscillation and the energy supersolution use the pointwise structure of the Euler nonlinearity. Whether a bounded ancient pressureless flow in three dimensions is constant is stated above as a conjecture.

References

  1. Arnold, V. I. (1966). Sur la géométrie différentielle des groupes de Lie de dimension infinie et ses applications à l'hydrodynamique des fluides parfaits. Annales de l'Institut Fourier, 16(1), 319–361. DOI
  2. Ebin, D. G. and Marsden, J. E. (1970). Groups of diffeomorphisms and the motion of an incompressible fluid. Annals of Mathematics, 92(1), 102–163. DOI
  3. Rouchon, P. (1992). The Jacobi equation, Riemannian curvature and the motion of a perfect incompressible fluid. European Journal of Mechanics B/Fluids, 11(3), 317–336.
  4. Misiołek, G. (1993). Stability of flows of ideal fluids and the geometry of the group of diffeomorphisms. Indiana University Mathematics Journal, 42(1), 215–235. DOI
  5. Preston, S. C. (2004). For ideal fluids, Eulerian and Lagrangian instabilities are equivalent. Geometric and Functional Analysis, 14(5), 1044–1064.
  6. Vieillefosse, P. (1982). Local interaction between vorticity and shear in a perfect incompressible fluid. Journal de Physique, 43(6), 837–842. DOI
  7. Cantwell, B. J. (1992). Exact solution of a restricted Euler equation for the velocity gradient tensor. Physics of Fluids A, 4(4), 782–793. DOI
  8. Leray, J. (1934). Sur le mouvement d'un liquide visqueux emplissant l'espace. Acta Mathematica, 63, 193–248. DOI
  9. Hopf, E. (1951). Über die Anfangswertaufgabe für die hydrodynamischen Grundgleichungen. Mathematische Nachrichten, 4(1–6), 213–231. DOI
  10. Scheffer, V. (1976). Partial regularity of solutions to the Navier–Stokes equations. Pacific Journal of Mathematics, 66(2), 535–552. DOI
  11. Caffarelli, L., Kohn, R., and Nirenberg, L. (1982). Partial regularity of suitable weak solutions of the Navier–Stokes equations. Communications on Pure and Applied Mathematics, 35(6), 771–831. DOI
  12. Seregin, G. and Šverák, V. (2009). On type I singularities of the local axi-symmetric solutions of the Navier–Stokes equations. Communications in Partial Differential Equations, 34(2), 171–201. DOI
  13. Escauriaza, L., Seregin, G. A., and Šverák, V. (2003). -solutions of the Navier–Stokes equations and backward uniqueness. Russian Mathematical Surveys, 58(2), 211–250. DOI
  14. Seregin, G., Silvestre, L., Šverák, V., and Zlatoš, A. (2012). On divergence-free drifts. Journal of Differential Equations, 252(1), 505–540. DOI
  15. Tao, T. (2016). Finite time blowup for an averaged three-dimensional Navier–Stokes equation. Journal of the American Mathematical Society, 29(3), 601–674. DOI
  16. Khesin, B., Langøen, R., and Markina, I. (2025). Curvatures of measure-preserving diffeomorphism groups of non-orientable surfaces. Nonlinearity, 38(5), 055017. DOI
  17. Ladyzhenskaya, O. A. and Seregin, G. A. (1999). On partial regularity of suitable weak solutions to the three-dimensional Navier–Stokes equations. Journal of Mathematical Fluid Mechanics, 1(4), 356–387. DOI
  18. Koch, G., Nadirashvili, N., Seregin, G., and Šverák, V. (2009). Liouville theorems for the Navier–Stokes equations and applications. Acta Mathematica, 203(1), 83–105. DOI