Abstract
In this paper, we prove that any Ricci shrinker that is sufficiently close to $(\mathbb C P^N,g_{FS})$ in the Gromov-Hausdorff sense must itself be isometric to $(\mathbb C P^N,g_{FS})$.
1. Introduction
The classification of Ricci shrinkers is a fundamental problem in the study of Ricci flow singularities. Ricci shrinkers represent the self-similar solutions that model the "limit" shapes of a manifold as it approaches a singularity. While complete classifications exist in dimensions 2 and 3, higher-dimensional shrinkers remain largely unclassified. A central question in this field is the rigidity of known examples, such as the complex projective space $(\mathbb{C}P^N, g_{FS})$. Rigidity implies that the space is isolated within the space of all Ricci shrinkers; no small deformation of the metric can produce a distinct, non-isometric shrinker.
2. Background: Ricci Flow and Shrinkers
In differential geometry, the Ricci flow deforms a manifold's metric to smooth out curvature, similar to the diffusion of heat. This process can develop singularities where curvature becomes infinite. Ricci shrinkers are the essential mathematical objects used to study these singularities. A Ricci shrinker is a Riemannian manifold $(M, g)$ satisfying the equation $Rc + \nabla^2 f = \frac{1}{2}g$ for some smooth function $f$.
Previous work had proven rigidity for even complex dimensions ($N=2m$), but the case for odd complex dimensions ($N=2m-1$) remained an open challenge. This paper addresses this gap, providing a unified proof for all $N$.
3. Methodology: Deformation and Obstruction Theory
The authors employ a combination of geometric analysis and deformation theory to prove rigidity. The approach follows three primary logical steps:
3.1 Weak-Compactness and Diffeomorphism
Using established weak-compactness theory for Ricci shrinkers, the authors show that any shrinker sufficiently close to $\mathbb{C}P^N$ in the Gromov-Hausdorff sense must be smooth and diffeomorphic to $\mathbb{C}P^N$. This allows the analysis to proceed by examining metrics on a fixed manifold.
3.2 Infinitesimal Deformations
The authors analyze the Ricci shrinker operator $\Phi(g)$. If a space is not rigid, there should be an infinitesimal deformation—a "direction" in the space of metrics—that allows for a change while remaining a Ricci shrinker. This involves solving the linearized shrinker equation.
3.3 High-Order Obstruction Theory
The core technical contribution of the paper is the use of obstruction theory. Even if a deformation exists at the first order (linearized level), it might be "blocked" at higher orders. The authors calculate the second-order and third-order variations of the Ricci shrinker equation. For $\mathbb{C}P^{2m-1}$, a second-order candidate for deformation exists, necessitating a move to the third order. By proving the third-order variation $I(h)$ is non-zero (Theorem 3.15), the authors demonstrate that no such path of shrinkers can exist. This distinction between an infinitesimal deformation (tangent vector) and an actual deformation (path) is critical to the proof of non-integrability.
4. Key Results
4.1 Main Rigidity Theorem
Theorem 1.1: For any $N \geq 1$, there exists a constant $\epsilon > 0$ such that any Ricci shrinker within a Gromov-Hausdorff distance $\epsilon$ of $(\mathbb{C}P^N, g_{FS})$ is isometric to it.
4.2 The Odd-Even Divide
The paper highlights a subtle difference between $\mathbb{C}P^{2m}$ and $\mathbb{C}P^{2m-1}$. In even dimensions, deformations are blocked at the second order. In odd dimensions, the authors successfully identified the third-order obstruction, proving that the infinitesimal deformation is "not integrable."
4.3 Rigidity of Product Spaces
The authors extend the discussion to product spaces, proving that the product of a complex projective space and certain Einstein manifolds (e.g., $\mathbb{C}P^N \times M_2$) is also rigid (Theorem 4.6).
5. Discussion and Future Directions
While this result establishes local rigidity, it does not rule out the existence of geometrically distant Ricci shrinkers with similar topological features. The authors note that the proof is more straightforward if the manifold is assumed to be Kähler from the start; a broader challenge remains in classifying shrinkers without such restrictive symmetry assumptions.
Furthermore, the authors propose two main conjectures for future work:
- General Product Rigidity: The product of any two complex projective spaces is rigid (Conjecture 4.7).
- Non-compact Extensions: A similar rigidity holds for certain non-compact Ricci shrinkers, such as the product of $\mathbb{C}P^N$ and flat Euclidean space $\mathbb{R}^k$ (Conjecture 4.8).
This work provides a robust framework for understanding the isolation of symmetric spaces in the landscape of singularity models, particularly through the rigorous application of high-order variational analysis.
Summary updated