2025-11-18T18:28:14.076279

Sharp Spectral Gap Estimates on Manifolds under Integral Ricci Curvature Bounds

Olivé, Seto, Tuerkoen
We prove sharp spectral gap estimates on compact manifolds with integral curvature bounds. We generalize the results of Kröger (Kröger '92) as well as of Bakry and Qian (Bakry-Qian '00) to the case of integral curvature and confirm the conjecture in (Ramos et al. '20) for the case $n \geq 3$.
academic

Sharp Spectral Gap Estimates on Manifolds under Integral Ricci Curvature Bounds

Basic Information

  • Paper ID: 2510.27083
  • Title: Sharp Spectral Gap Estimates on Manifolds under Integral Ricci Curvature Bounds
  • Authors: Xavier Ramos Olivé, Shoo Seto, Malik Tuerkoen
  • Classification: math.DG (Differential Geometry), math.AP (Analysis and Partial Differential Equations), math.SP (Spectral Theory)
  • Publication Date: October 31, 2025
  • Paper Link: https://arxiv.org/abs/2510.27083

Abstract

This paper establishes sharp spectral gap estimates on compact manifolds with integral curvature bounds. The authors generalize results of Kröger (1992) and Bakry-Qian (2000) to the integral curvature setting and confirm the conjecture of Ramos et al. (2020) for dimension n≥3n \geq 3.

Research Background and Motivation

Problem Statement

This paper investigates lower bounds for the first nonzero eigenvalue λ1\lambda_1 of the Laplacian on compact Riemannian manifolds, particularly under integral Ricci curvature bounds.

Significance of the Problem

  1. Core Problem in Geometric Analysis: Eigenvalue estimates represent a classical problem in geometric analysis, connecting geometric properties of manifolds (such as curvature and diameter) with spectral properties.
  2. Generalization of Classical Results:
    • Lichnerowicz (1958) proved estimates under pointwise positive Ricci curvature bounds
    • Zhong-Yang (1984) established λ1≥π2/D2\lambda_1 \geq \pi^2/D^2 under non-negative Ricci curvature
    • Bakry-Qian (2000) and Kröger (1992) unified these results into a single framework
  3. Unified Framework: The classical results are unified through a one-dimensional model approach.

Limitations of Existing Methods

  1. Pointwise Curvature Assumptions Too Strong: Classical results require pointwise Ricci curvature bounds Ric≥(n−1)K\text{Ric} \geq (n-1)K.
  2. Integral Curvature Conditions More Natural: In many geometric problems, integral curvature conditions are weaker and more natural than pointwise conditions.
  3. Existing Integral Curvature Results Not Sharp: While Aubry (2007) and Ramos et al. (2020) considered integral curvature, they did not achieve sharp estimates.

Research Motivation

Ramos et al. (2020) explicitly conjectured that sharp estimates in the integral curvature setting should be obtainable using auxiliary function methods. This paper confirms this conjecture for n≥3n \geq 3.

Core Contributions

  1. Main Theorem (Theorem 1.3): Under integral Ricci curvature conditions k(p,K)<ε0k(p,K) < \varepsilon_0 (where p>n/2p > n/2), for any α∈(0,1)\alpha \in (0,1): λ1(M)≥αλ1(n,K,D)\lambda_1(M) \geq \alpha\lambda_1(n,K,D) where λ1(n,K,D)\lambda_1(n,K,D) is the first nonzero eigenvalue of the one-dimensional model problem.
  2. Sharpness: As k(p,K)→0k(p,K) \to 0, the estimate approaches the classical pointwise curvature case, recovering the Bakry-Qian and Kröger results.
  3. Gradient Comparison Theorem (Theorem 3.1): Establishes gradient comparison estimates for eigenfunctions under integral curvature conditions: J∣∇u∣2≤(w′)2∘w−1(u)J|\nabla u|^2 \leq (w')^2 \circ w^{-1}(u)
  4. Conjecture Confirmation: Confirms the conjecture from reference 10 for dimension n≥3n \geq 3.
  5. Technical Innovation: Introduces an auxiliary function JJ to absorb integral curvature terms, overcoming the difficulty of lacking pointwise curvature bounds.

Detailed Methodology

Problem Setup

Given an nn-dimensional compact Riemannian manifold MM (possibly with convex C2C^2 boundary), consider the eigenvalue problem for the Laplacian: Δu+λu=0 in M\Delta u + \lambda u = 0 \text{ in } M with Neumann boundary conditions when the boundary is nonempty. The goal is to estimate the first nonzero eigenvalue λ1\lambda_1 under integral Ricci curvature conditions.

Core Concepts

Integral Curvature Metric: Define ρK(x)=max⁡{−(ρ(x)−(n−1)K),0}\rho_K(x) = \max\{-(\rho(x) - (n-1)K), 0\} where ρ(x)\rho(x) is the minimum eigenvalue of the Ricci tensor at xx. The integral curvature condition is: k(p,K)=(1Vol(M)∫MρKpdV)1/pk(p,K) = \left(\frac{1}{\text{Vol}(M)}\int_M \rho_K^p dV\right)^{1/p}

One-Dimensional Model: Consider the one-dimensional eigenvalue problem:

w'' - T_{n,K}w' + \lambda w = 0 & \text{on } [-D/2, D/2]\\ w'(\pm D/2) = 0 \end{cases}$$ where $$T_{n,K}(x) = \begin{cases} -(n-1)\sqrt{K}\tan(\sqrt{K}x) & K > 0\\ 0 & K = 0\\ -(n-1)\sqrt{-K}\tanh(\sqrt{-K}x) & K < 0 \end{cases}$$ ### Method Architecture The proof strategy consists of three key steps: #### 1. Auxiliary Function Method (Section 2) **Definition 2.4**: Introduce a positive function $J$ satisfying: $$\Delta J - \tau\frac{|\nabla J|^2}{J} - 2J\rho_K = -\sigma J$$ where $\tau > 1$, $\sigma \geq 0$. **Key Role of Lemma 2.5**: For any $\delta > 0$, there exists $\varepsilon = \varepsilon(n,p,D,\tau)$ such that when $k(p,K) \leq \varepsilon$, there exist $\sigma$ and corresponding $J$ satisfying $0 \leq \sigma \leq 4\varepsilon$ and $|J-1| \leq \delta$. The clever design of this auxiliary function: - Through the transformation $J = W^{-1/(\tau-1)}$, the equation becomes equivalent to the eigenvalue equation $\Delta W + VW = \tilde{\sigma}W$, where $V = 2(\tau-1)\rho_K$ - Using Poincaré and Sobolev inequalities, one can control $W$; these tools under integral curvature conditions were established by Gallot et al. #### 2. Gradient Comparison (Section 3) **Proof Strategy of Theorem 3.1**: 1. **Parametric Perturbation**: Introduce perturbation parameters $\tilde{K} < K$, $N > n$, $\lambda = (1+2\delta)\lambda_1$, such that the one-dimensional model coefficients $T$ satisfy the Riccati equation: $$T' = \frac{T^2}{N-1} + (N-1)\tilde{K}$$ 2. **Proof by Contradiction**: Assume there exists a point $x$ where $J|\nabla u|^2 > (w')^2(w^{-1}(u(x)))$. Define: $$Q = J|\nabla u|^2 - (cw')^2((cw)^{-1}(u))$$ Choose $c > 1$ such that $Q(x) = 0$ at the maximum point $x$. 3. **Maximum Principle Calculation**: At the maximum point $x$: - $\nabla Q(x) = 0$ (first-order condition) - $\Delta Q(x) \leq 0$ (second-order condition) 4. **Bochner Formula Application**: Use: $$\frac{1}{2}\Delta|\nabla u|^2 = |\text{Hess}u|^2 + \text{Ric}(\nabla u, \nabla u) + \langle\nabla u, \nabla\Delta u\rangle$$ and the refined Cauchy-Schwarz inequality: $$|\text{Hess}u|^2 \geq \frac{\lambda_1^2 u^2}{n-1} + \frac{n}{n-1}A_u^2 + \frac{2\lambda_1 u}{n-1}A_u$$ 5. **Key Inequality**: After detailed calculations, obtain (equation (25)): $$0 \geq J(1-\alpha)\frac{\lambda_1^2 c^2w^2}{n-1} + \frac{n}{n-1}(1-\beta)\frac{c^2(w'')^2}{J} + ((n-1)\tilde{K} - \lambda_1 - \sigma)(cw')^2 - \text{other terms}$$ 6. **Parameter Selection**: By carefully choosing $\alpha, \beta, N, \tilde{K}$, ensure all coefficients are non-negative, leading to a contradiction. **Technical Difficulties**: - Must simultaneously handle upper and lower bound constraints on $y = J\lambda_1/\lambda$ (inequalities (26)-(28)) - For $n = 2$, the method fails (root analysis shows inability to obtain sharp estimates) #### 3. Extremal Matching (Section 4) **Objective of Theorem 4.5**: Prove that there exists an eigenfunction $w$ of the one-dimensional model whose domain $[a,b]$ and extremal values satisfy: $$u_* = \max_I w, \quad -1 = \min u = \min_I w$$ **Proof Strategy**: 1. **Maximum Value Comparison (Theorem 4.1)**: Prove by contradiction that $u_* \geq m_{N,\tilde{K}}$ (maximum of the one-dimensional model): - Use Proposition 4.2 to establish measure comparison - Apply Proposition 4.3 for integral estimates - Lemma 4.4 provides volume lower bounds, ultimately deriving a contradiction 2. **Continuity Argument**: Use the continuous dependence of the one-dimensional model eigenfunction's maximum on the starting point $a$ to find suitable $a$ matching the extremal values. 3. **Case Analysis**: - $K > 0$: Use Aubry estimates and domain monotonicity - $K < 0$: Further divide into $\lambda \leq -(N-1)^2\tilde{K}/4$ and $\lambda > -(N-1)^2\tilde{K}/4$ ### Technical Innovations 1. **Auxiliary Function Absorbs Integral Curvature**: The design of $J$ allows the integral curvature term $\rho_K$ to be "absorbed" through $J$'s equation in maximum principle calculations. 2. **Parametric Perturbation Technique**: By introducing perturbation parameters $N > n$, $\tilde{K} < K$, $\lambda > \lambda_1$, recover sharp estimates in the limit. 3. **Application of Zhang-Zhu Technique**: Adapt techniques from reference [14] for handling integral curvature conditions. 4. **Continuity Approximation**: Use the continuous dependence of $\lambda_1(n,K,D)$ on parameters to obtain sharp estimates through approximation. ## Experimental Setup This is a pure theoretical mathematics paper with no numerical experiments or datasets. All results are obtained through rigorous mathematical proofs. ## Experimental Results ### Main Theoretical Results **Statement of Theorem 1.3**: Let $M$ be an $n$-dimensional compact Riemannian manifold ($n \geq 3$) with diameter $D$ and $p > n/2$. For any $\alpha \in (0,1)$, there exists $\varepsilon_0 = \varepsilon_0(n,p) > 0$ such that when $k(p,K) < \varepsilon_0$: $$\lambda_1(M) \geq \alpha\lambda_1(n,K,D)$$ **Sharpness**: When $k(p,K) = 0$ (i.e., pointwise curvature condition $\text{Ric} \geq (n-1)K$), recover $\lambda_1(M) \geq \lambda_1(n,K,D)$. **Corollary for Negative Curvature**: For $K < 0$, under the conditions of Theorem 1.3, for any $\alpha \in (0,1)$: $$\lambda_1 \geq \alpha\frac{\pi^2}{D^2}\exp(-c_nD\sqrt{(n-1)|K|})$$ where $c_n = \max\{2, n-1\}$, which is the integral curvature version of Yang's (1990) estimate. ### Comparison with Existing Results 1. **Generalization of Lichnerowicz Estimate**: - Classical: $\text{Ric} \geq (n-1)K > 0 \Rightarrow \lambda_1 \geq nK$ - Aubry (2007): $\lambda_1 \geq nK(1 - C(n,p)k(p,K))$ - This paper: Recovers the sharp estimate of the one-dimensional model 2. **Generalization of Zhong-Yang Estimate**: - Classical: $\text{Ric} \geq 0 \Rightarrow \lambda_1 \geq \pi^2/D^2$ - Ramos et al. (2020): Estimates under integral curvature conditions - This paper: Sharp estimates in a unified framework 3. **Generalization of Bakry-Qian Unified Theorem**: - Classical: Unified estimates under pointwise curvature - This paper: Unified estimates under integral curvature ### Theoretical Findings 1. **Dimension Restriction**: The method is effective for $n \geq 3$, but fails for $n = 2$ (see analysis of equation (26) in the proof). 2. **Integral Exponent Restriction**: Requires $p > n/2$, related to critical exponents in Sobolev embedding. 3. **Parameter Dependence**: - $\varepsilon_0$ depends on $n, p, D, \tau$ - Explicit construction of perturbation parameters $N, \tilde{K}$ in formulas (29)-(31) ## Related Work ### Classical Eigenvalue Estimates 1. **Lichnerowicz (1958)**: First established eigenvalue lower bounds under positive Ricci curvature: $$\lambda_1 \geq \frac{n}{n-1}\inf \text{Ric}$$ 2. **Zhong-Yang (1984)**: Non-negative Ricci curvature case: $$\lambda_1 \geq \frac{\pi^2}{D^2}$$ 3. **Yang (1990)**: Exponential decay estimates for negative curvature. ### Unified Frameworks 1. **Kröger (1992)**: Introduced the one-dimensional model method, unifying various estimates through gradient comparison. 2. **Bakry-Qian (2000)**: Established more general frameworks on CD(R,n) spaces (metric measure spaces satisfying curvature-dimension conditions). 3. **Valtorta (2012), Naber-Valtorta (2014)**: Extensions to $p$-Laplacian cases. ### Integral Curvature Conditions 1. **Gallot (1988)**: Established isoperimetric inequalities under integral curvature conditions. 2. **Petersen-Wei (1997), Petersen-Sprouse (1998)**: Volume comparison under integral curvature. 3. **Dai-Wei-Zhang (2018)**: Local Sobolev constant estimates. 4. **Aubry (2007)**: Lichnerowicz estimates under integral curvature. 5. **Ramos et al. (2020)**: Zhong-Yang estimates under integral curvature, proposing the conjecture confirmed in this paper. ### Auxiliary Function Methods **Zhang-Zhu (2017)**: Introduced auxiliary function techniques in Li-Yau gradient estimates under integral curvature conditions, which this paper adapts and develops. ## Conclusions and Discussion ### Main Conclusions 1. Established sharp spectral gap estimates under integral Ricci curvature conditions, generalizing classical results of Kröger and Bakry-Qian. 2. Confirmed the conjecture of Ramos et al. (2020) for $n \geq 3$. 3. Proved that as integral curvature approaches zero, estimates converge to the sharp estimates of the pointwise curvature case. ### Limitations 1. **Dimension Restriction**: The method is effective only for $n \geq 3$. The paper explicitly notes that the $n = 2$ case cannot be handled by this method (see discussion in Theorem 3.1 proof). 2. **Integral Exponent Requirement**: Requires $p > n/2$, a technical limitation. 3. **Small Curvature Assumption**: Requires $k(p,K) < \varepsilon_0$, i.e., integral curvature must be sufficiently small. 4. **Non-Explicit Constants**: While parameters $N, \tilde{K}$ have explicit formulas, the specific value of $\varepsilon_0$ is not explicit. ### Future Directions 1. **The $n=2$ Case**: Seek new methods to handle the two-dimensional case. 2. **Optimization of Integral Exponent**: Can the restriction $p > n/2$ be relaxed? 3. **Large Curvature Regime**: Study cases where $k(p,K)$ is not small. 4. **Other Operators**: Extend to $p$-Laplacian or more general elliptic operators. 5. **Applications**: Apply the method to concrete geometric problems. ## In-Depth Evaluation ### Strengths 1. **Significant Theoretical Contribution**: - Resolves an open conjecture in the field - Generalizes classical results to the more general integral curvature setting - Establishes sharp estimates that recover known optimal results in the limit 2. **Strong Technical Innovation**: - Clever application of auxiliary function methods - Fine control of parametric perturbation techniques - Successful adaptation of Zhang-Zhu techniques to eigenvalue estimation problems 3. **Rigorous and Complete Proofs**: - Clear logical structure with three main steps (gradient comparison, extremal matching, diameter comparison) - Detailed discussion of different cases (sign of $K$, boundary conditions, etc.) - Careful treatment of technical details (e.g., Lemma 3.3 handling boundary cases) 4. **Clear Exposition**: - Well-organized structure with clear objectives for each section - Clear definitions of key concepts - Adequate explanation of main ideas and technical difficulties ### Weaknesses 1. **Dimension Restriction**: - The $n = 2$ case remains unresolved, with authors noting current methods cannot handle it - This limits the generality of the results 2. **Non-Constructive Nature**: - While existence is proven, the specific value of $\varepsilon_0$ is unclear - This may cause difficulties in applications 3. **Small Curvature Assumption**: - Requires $k(p,K)$ to be sufficiently small, but the specific range is unclear - Quality of estimates in the large curvature regime is unknown 4. **Technical Complexity**: - Proof involves fine control of multiple perturbation parameters - Complex interdependencies among parameters ($\alpha, \beta, N, \tilde{K}, \delta$) ### Impact 1. **Theoretical Contribution**: - Provides fundamental tools for spectral theory under integral curvature conditions - Confirms important conjecture, advancing the field - Methods may inspire generalizations of other geometric inequalities to integral curvature settings 2. **Potential Applications**: - Applicable to geometric flows, harmonic analysis, and related fields - Provides tools for studying manifolds "almost satisfying" curvature conditions 3. **Methodological Value**: - Auxiliary function techniques may apply to other problems - Parametric perturbation and continuity approximation strategies have general applicability 4. **Open Problems**: - The $n = 2$ case becomes a new research direction - Stimulates further understanding of the nature of integral curvature conditions ### Applicable Scenarios 1. **Theoretical Research**: - Study spectral properties of manifolds under integral curvature conditions - Establish connections between geometry and analysis 2. **Geometric Analysis**: - Analyze manifolds satisfying weak curvature conditions - Study effects of curvature perturbations on spectra 3. **Application Fields**: - Spectral estimates in geometric flows - Energy level estimates in quantum mechanics (Laplacian corresponds to Hamiltonian) 4. **Limitations**: - Not applicable to low dimensions ($n=2$) - Requires integral curvature to be sufficiently small ## References The paper cites 15 key references, including: 1. **[2] Bakry-Qian (2000)**: One of the main objects of generalization in this paper 2. **[5] Kröger (1992)**: Pioneering work on gradient comparison methods 3. **[6] Lichnerowicz (1958)**: Foundational work on eigenvalue estimates 4. **[10] Ramos et al. (2020)**: Proposed the conjecture confirmed in this paper 5. **[14] Zhang-Zhu (2017)**: Source of auxiliary function methods 6. **[4] Gallot (1988), [8] Petersen-Sprouse (1998)**: Foundational tools for integral curvature conditions --- **Overall Assessment**: This is a high-quality pure mathematics paper that makes significant progress in spectral estimation theory under integral curvature conditions. Through clever auxiliary function methods and fine technical control, the authors successfully generalize classical sharp spectral estimates to the integral curvature setting and confirm an important conjecture in the field. Despite limitations such as dimension restrictions, the paper's theoretical contributions are substantial, the methods are innovative, and it has significant impact on the field of geometric analysis.