A finite Livshits theorem and local length spectrum rigidity

summary

The gist

A finite Livshits theorem and local length spectrum rigidity shows that for closed negatively curved Riemannian manifolds whose length spectrum is exponentially separated…

In short

The paper proves that for closed negatively curved manifolds with exponentially separated length spectra, these spectra are local isometry invariants. It achieves this by combining a finite Livshits theorem with stability estimates on the X-ray transform, showing that if two metrics share the same length spectrum and are close in a certain topology, they must be isometric.

Key concepts

Local Length Spectrum Rigidity
This concept shows that knowing the lengths of all closed geodesics (the length spectrum) is enough to locally determine the Riemannian metric up to isometry. If two metrics have the same spectrum and are close enough in a specific topology, they must be essentially the same metric.
Finitary Approximate Livshits Theorem
This theorem provides a quantitative version of spectral rigidity, showing that if two metrics are close, the error in matching their length spectra is exponentially small. It uses advanced machinery to ensure this stability holds even for higher regularity of the metrics.
X-ray Transform Stability
The proof strategy relies on showing that if two metrics have the same length spectrum and their solenoidal parts are small, they must be isometric. This involves using an injective X-ray transform operator combined with stability estimates developed by key authors to link spectral data to geometric closeness.

Terminology used across episodes

This episode discusses

The paper

A finite Livshits theorem and local length spectrum rigidity · Read on arXiv

JONATHAN DEWITT, SPENCER DURHAM, JAMES MARSHALL REBER, THOMAS ALOYSIUS O’HARE

National Science Foundation

Transcript

Introduction to the show: ident: Mathematics. The day in mathematics: new theorems, new proofs, and what they open.

Ada: I'm Ada.

Emmy: And I'm Emmy, here with Kolya.

Ada: We're looking at the paper: "A finite Livshits theorem and local length spectrum rigidity".

Emmy: Here's the gist of the paper. A finite Livshits theorem and local length spectrum rigidity shows that for closed negatively curved Riemannian manifolds whose length spectrum is exponentially separated, the length spectrum is a local isometry invariant.

Ada: Local Length Spectrum Rigidity The main result in this paper is to show that for nice metrics, the length spectrum locally determines the metric up to isometry in an appropriately fine Ck-topology.

Emmy: Theorem A states that if (M, g0) is a closed, connected C∞ Riemannian manifold with everywhere negative sectional curvature whose length spectrum is exponentially separated, there exists k ∈ N and ε > 0 such that if g is a Ck Riemannian metric satisfying LS(g) = LS(g0) and∥g0 − g∥Ck(M) < ε, then there is a Ck diffeomorphism φ: M → M isotopic to the identity such that g0 = φ∗(g).

Ada: Corollary A further states that if (M, g0) is a smooth, closed, connected hyperbolic manifold with dimension at least three, then there exists a C∞-neighborhood U of g0 in NegRiem∞(M) such that if g ∈ U satisfies LS(g0) = LS(g), then g0 and g are isometric.

Paper summary: Ada: So we're looking at this paper by DeWitt, Durham, Reber, and O’Hare called "A finite Livshits theorem and local length spectrum rigidity." The main idea here is that for closed negatively curved Riemannian manifolds where the length spectrum is exponentially separated, that length spectrum acts like a local isometry invariant.

Emmy: It claims that if you have two metrics with the same length spectrum, and one metric is very close to the other in a certain way, then they must be isometric up to some small perturbation. This holds for a dense set of negatively curved metrics, which is pretty significant because it means this rigidity isn't just happening on one specific example.

Kolya: What I find interesting is that the paper uses a finitary version of the Livshits theorem with an exponential tail, which seems to be the engine driving this result. The abstract mentions that they use stability machinery from Guillarmou, Lefeuvre, and Gou¨ezel along with dynamical cohomology.

Ada: Exactly. They are using that machinery to solve a cohomological equation—specifically the R-valued Livshits theorem—to produce an invariant structure or show how one map intertwines two structures. It's essentially showing that certain functions are coboundaries, which tells you something fundamental about the geometry of the manifold itself.

Emmy: And this leads to Theorem A, which states that for a smooth, closed, connected hyperbolic manifold with dimension at least three, if you have a metric g0 and another metric g close to it in the C infinity topology where their length spectra are equal, then they must be isometric.

Kolya: That's a strong claim because it requires the length spectrum to be exponentially separated for that conclusion to hold. I wonder what happens if we look at cases where the separation isn't exponential, or if we consider lower dimensions?

Ada: The paper specifically points out that for hyperbolic metrics, this exponential separation is known when the dimension of M is at least three because of earlier work by DeWitt and others. It also notes that this result holds for a dense set of negatively curved metrics, which makes it much broader than just one particular metric g0.

Emmy: The finitary approximate Livshits theorem mentioned in the paper is what allows them to get a quantitative version of this idea, suggesting that there's an exponentially small error when the separation parameter epsilon is zero and holding for higher regularity C k alpha.

Kolya: So, if we translate this to something more practical, it means that even when things are complicated, like with higher regularity C k alpha metrics, the rigidity still holds with a small error term that shrinks exponentially. That's a nice quantitative statement compared to just saying "they are close."

Paper summary: Ada: It gets more involved when they talk about the proof strategy. They combine this finitary Livshits theorem with stability estimates on what's called the X-ray transform operator. The key is showing that if two metrics share the same length spectrum and the solenoidal part of their difference is small, then it must be zero.

Emmy: The construction of this proof relies on an injective X-ray transform operator combined with exponential separation and stability estimates developed by Gou¨ezel, Guillarmou, and Lefeuvre. It’s a sophisticated way to connect spectral information to geometric structure.

Kolya: And the paper also has this quantitative iterative scheme, Theorem four point one.one which lets them construct a global KAM scheme where they don't have to assume that either the C k alpha norm or the C zero norm are small when starting out <ref:2609.23762#pg1>. It builds up a sequence of functions that are cohomologous to the initial function in an inductive way until it hits a point where the conditions of their lemma fail.

Ada: That iterative process is crucial for managing the complexity of these functional equations across different regularity levels, which is where they handle those higher C k alpha norms. Then, in the final step, they apply this finitary Livshits theorem to a specific function related to the difference between g0 and g.

Emmy: They use an estimate involving a quantity called T, which is defined using the log of C times the inverse of some expression involving the norm of that difference. This lets them conclude that if there's a coincidence in length spectrum, it implies a coincidence in what they call the marked length spectrum for geodesics up to a certain length.

Kolya: So, what this means for someone who just listens to our show is that if you see two metrics with the same set of closed geodesic lengths, and they are close enough, the paper suggests that those two metrics are essentially isometric in a very fine sense. It's about how the geometry dictates the spectral data.

Ada: It really boils down to this final conclusion: if k is large enough, and you choose epsilon k alpha small enough based on that bound, then any equality you see in their setup can only happen if the metrics are already isometric and C k alpha isometric.

Emmy: The paper "A finite Livshits theorem and local length spectrum rigidity" suggests that for negatively curved manifolds with exponentially separated length spectra, the length spectrum is indeed a local isometry invariant, which is a powerful statement about how geometry determines its own spectral data.

Kolya: It implies that you don't need to know every single piece of the metric to determine its shape locally; just knowing the spectrum gives you enough information if the separation condition is met.

Ada: That's right. The paper uses a finitary Livshits theorem with an exponential tail as the main tool, and it shows that this allows us to rigorously prove this local determination of the metric up to isometry in a suitable topology.

Conclusion: Ada: So we're wrapping up this paper, "A finite Livshits theorem and local length spectrum rigidity." Basically, they're showing that if you know the set of closed geodesic lengths for a negatively curved manifold, you can locally figure out what the metric actually is.

Emmy: It’s about taking spectral data—the length spectrum—and using it to pinpoint the geometry itself up to some small error term. It suggests that this length information is a very strong fingerprint for the local shape of the space.

Kolya: What I find interesting here is how they get this rigidity down to a quantitative statement, dealing with those exponential separations and these C k alpha norms. It’s not just a qualitative statement about closeness; it’s about how fast that error term shrinks as you go to higher regularity.

Ada: Right, and the authors use a finitary version of the Livshits theorem—that's that core tool they build their argument on—to nail down these properties. It shows that this method works even with those higher C k alpha requirements.

Emmy: The authors are connecting dynamical systems theory, like Anosov flows, with spectral geometry in a really deep way here. They use things like the X-ray transform operator and stability estimates to bridge that gap between the flow dynamics and the metric structure.

Kolya: So, for someone just listening to the show, what does this mean? It means that if you have two negatively curved spaces with exponentially separated lengths, they are essentially isometric locally. That’s a pretty concrete geometric statement about how much information is needed to describe a space.

Ada: Exactly. The implication is that for these specific types of manifolds, the length spectrum is not just some random set of numbers; it carries enough structure to determine the metric itself in a local sense.

Emmy: It opens up new ways to study these systems because now we have a spectral way to probe the underlying geometry more deeply than before. It’s about using dynamics as a tool for inverse problems in geometry.

Kolya: And what's next? They seem to be setting the stage for even more quantitative results, like how much smaller that error term can actually be when you move from one regularity level to another. We'll see how that plays out in the next piece of work.

More episodes

← Home