Self-intersection of the relative dualizing sheaf on modular curves $X_1(N)$

Basic Information

• Paper ID: 1212.1294
• Title: Self-intersection of the relative dualizing sheaf on modular curves $X_1(N)$
• Author: Hartwig Mayer
• Classification: math.NT (Number Theory)
• Submission Date: Submitted to arXiv on December 6, 2012; updated on October 22, 2018
• Paper Link: https://arxiv.org/abs/1212.1294

Abstract

This paper investigates the stable arithmetic self-intersection number of the relative dualizing sheaf on modular curves $X_1(N)/\mathbb{Q}$. For $N$ an odd, square-free positive integer divisible by at least two coprime integers greater than or equal to 4, the author establishes an asymptotic formula for the stable arithmetic self-intersection number of the relative dualizing sheaf, expressed entirely in terms of $N$. Based on the main theorem, the author also derives an asymptotic formula for the stable Faltings height of the Jacobian $J_1(N)/\mathbb{Q}$, and an effective version of the Bogomolov conjecture on $X_1(N)/\mathbb{Q}$ for sufficiently large $N$.

Research Background and Motivation

1. Core Problem: Computing the stable arithmetic self-intersection number $\omega_N^2$ of the relative dualizing sheaf $\omega_{X_1(N)/\mathbb{Z}[\zeta_N]}$ on the modular curve $X_1(N)$.
2. Significance:
   • The strict positivity of the stable arithmetic self-intersection number is equivalent to the Bogomolov conjecture, which asserts that for curves of genus greater than 1, algebraic points embedded in the Jacobian are discretely distributed in the "Néron-Tate topology"
   • Important applications in computing Fourier coefficients of modular forms
   • Closely related to Faltings height theory
3. Limitations of Existing Research:
   • Known stable arithmetic self-intersection numbers are limited to genus 1 curves, genus 2 curves, and specific modular curves $X_0(N)$ (where $N$ is square-free and $2,3 \nmid N$)
   • Lack of precise asymptotic formulas for modular curves of type $X_1(N)$
4. Research Motivation: To fill the gap in the theory of arithmetic self-intersection numbers for $X_1(N)$ modular curves and provide tools for related problems in number theory and algebraic geometry.

Core Contributions

1. Main Theorem: For $N$ satisfying the stated conditions, the asymptotic formula holds: $\omega_N^2 = 3g_N \log(N) + o(g_N \log(N))$ where $g_N$ is the genus of $X_1(N)$.
2. Faltings Height Formula: An asymptotic formula for the stable Faltings height of the Jacobian $J_1(N)/\mathbb{Q}$ is obtained: $h_{\text{Fal}}(J_1(N)) = \frac{g_N}{4}\log(N) + o(g_N \log(N))$
3. Effective Version of the Bogomolov Conjecture: For sufficiently large $N$, an effective Bogomolov bound is established.
4. Technical Innovation: Spectral-theoretic methods applicable to congruence subgroups $\Gamma_1(N)$ are developed, extending techniques of Abbes-Ullmo and Jorgenson-Kramer.

Detailed Methodology

Task Definition

Computing the stable arithmetic self-intersection number $\omega_N^2 = \frac{1}{[\mathbb{Q}(\zeta_N):\mathbb{Q}]} \omega_{X_1(N)/\mathbb{Z}[\zeta_N]}^2$, where $\omega_{X_1(N)/\mathbb{Z}[\zeta_N]}$ is the relative dualizing sheaf.

Core Decomposition Formula

The author establishes a key decomposition formula (Proposition 7.6): $\omega_N^2 = 4g_N(g_N-1)g_{\text{can}}(0,\infty) + \frac{1}{\phi(N)}\frac{g_N+1}{g_N-1}(V_0,V_\infty)_{\text{fin}}$

where:

• $g_{\text{can}}(0,\infty)$ is the canonical Green function between cusps 0 and $\infty$
• $(V_0,V_\infty)_{\text{fin}}$ is the finite intersection number of vertical divisors
• $\phi(N)$ is Euler's totient function

Technical Architecture

1. Analytic Part Computation

Computing $4g_N(g_N-1)g_{\text{can}}(0,\infty)$ via spectral expansion theory:

Spectral Expansion: Utilizing the spectral expansion of automorphic kernel functions: $K_k(z,w) = \sum_{j=0}^{\infty} h(r_j)u_j(z)u_j(w) + \text{continuous spectrum terms}$

Fundamental Formula: Establishing the function: $F(z) := \frac{1}{g_N}\sum_{j=1}^{g_N} y^2|f_j(z)|^2$ where $\{f_j\}$ is an orthonormal basis of weight 2 cusp forms.

Rankin-Selberg Transform: Computing the Green function via the Laurent expansion at $s=1$ of: $R_F(s) = \int_{\Gamma_1(N)\backslash \mathbb{H}} F(z)E_{\infty,0}(z,s)\mu_{\text{hyp}}(z)$

2. Algebraic Part Computation

Computing the geometric intersection number $(V_0,V_\infty)_{\text{fin}}$, where $V_0, V_\infty$ are vertical divisors constructed via the fibral structure of the minimal regular model.

3. Asymptotic Analysis

Hyperbolic Contribution: Via the logarithmic derivative of the Selberg zeta function: $R_H(t,1) = -\frac{1}{2v_N}\int_0^t \Theta_{\Gamma_1(N)}(\xi)d\xi$

Parabolic and Spectral Contributions: Utilizing scattering matrix theory for Eisenstein series.

Technical Innovations

1. Extended Selberg Trace Formula: Generalizing the $X_0(N)$ case methodology to $X_1(N)$, requiring treatment of more complex congruence subgroup structures.
2. Improved Green Function Estimates: Combining work of Abbes-Ullmo and Jorgenson-Kramer to provide precise asymptotics for Green functions between cusps.
3. Explicit Computation of Geometric Intersection Numbers: Computing the fibral structure of the minimal regular model via Katz-Mazur theory.

Experimental Setup

Theoretical Verification Methods

This is pure mathematical theoretical research, verified primarily through:

1. Consistency Checks: Verifying consistency of asymptotic behavior across all components
2. Comparison with Known Cases: Comparing with known results for the $X_0(N)$ case by analogy
3. Numerical Estimates: Utilizing bounds on the Selberg zeta function to verify reasonableness of error terms

Technical Conditions

• $N$ is odd and square-free
• $N = N'qr$, where $q,r \geq 4$ and coprime
• $g_N \geq 1$ (i.e., $N = 11$ or $N \geq 13$)

Experimental Results

Main Results

Theorem 7.7 (Main Theorem): $\omega_N^2 = 3g_N \log(N) + o(g_N \log(N))$

Theorem 8.2 (Faltings Height): $h_{\text{Fal}}(J_1(N)) = \frac{g_N}{4}\log(N) + o(g_N \log(N))$

Asymptotic Analysis Results

1. Analytic Part: $4g_N(g_N-1)g_{\text{can}}(0,\infty) = 2g_N \log(N) + o(g_N \log(N))$
2. Algebraic Part: $(V_0,V_\infty)_{\text{fin}} = -\phi(N)\frac{24(g_N-1)^2}{\prod_{p|N}(p^2-1)}\sum_{p|N}\frac{p+1}{p-1}\log(p)$
3. Combined Result: The two parts contribute $2g_N \log(N)$ and $g_N \log(N)$ respectively, totaling $3g_N \log(N)$.

Effective Version of the Bogomolov Conjecture

Theorem 8.7: For sufficiently large $N$ and any $\varepsilon > 0$, the set of algebraic points $\{x \in X_1(N)(\mathbb{Q}) | h_{NT}(\phi_D(x)) < (\frac{3}{4} - \varepsilon)\log(N)\}$ is finite.

Related Work

Major Related Research

1. Foundations of Arakelov Theory: Arakelov (1974), Faltings (1984) established intersection theory on arithmetic surfaces
2. Modular Curves $X_0(N)$ Case: Pioneering work by Abbes-Ullmo (1997), Müller-Ullmo (1998)
3. Spectral Theory Methods: Automorphic form spectral theory by Iwaniec (2002), Roelcke (1966-1967)
4. Green Function Theory: Improvements in Green function bounds through work of Jorgenson-Kramer

Advantages of This Work Compared to Related Research

1. First Treatment of $X_1(N)$: Compared to known $X_0(N)$ results, the congruence subgroup structure of $X_1(N)$ is more complex
2. Complete Asymptotic Formula: Provides precise coefficients of the main term, rather than merely upper bounds
3. Unified Methodology: Organically combines analytic and algebraic-geometric methods

Conclusions and Discussion

Main Conclusions

1. A complete asymptotic theory for the self-intersection number of the relative dualizing sheaf on $X_1(N)$ modular curves is established
2. Effective versions of Faltings height and the Bogomolov conjecture are provided
3. Technical methods applicable to general congruence subgroups are developed

Limitations

1. Restrictive Conditions: Requires $N$ to be of specific form (odd, square-free, with specific factorization)
2. Asymptotic Nature: Only the main term is provided; the precise form of lower-order terms remains unclear
3. Computational Complexity: The methodology involves deep analytic number theory techniques, making practical computation difficult

Future Directions

1. Generalization to Arbitrary $N$: Handling cases with square factors or even $N$
2. Higher-Order Term Analysis: Determining the precise form of the $o(g_N \log N)$ term
3. Other Modular Curves: Extending the methodology to other modular curves such as $X(N), X_0^+(N)$

In-Depth Evaluation

Strengths

1. Technical Depth: Skillfully combines deep techniques from Arakelov geometry, spectral theory, and algebraic geometry
2. Completeness of Results: Not only provides the main theorem but also important arithmetic applications
3. Methodological Innovation: Successfully extends $X_0(N)$ techniques to the more difficult $X_1(N)$ case
4. Theoretical Significance: Makes important contributions to the arithmetic geometry theory of modular curves

Weaknesses

1. Readability: The technical content is extremely deep, requiring specialized background in multiple fields
2. Scope of Applicability: Strong restrictions on $N$ limit the practical range of applications
3. Computational Practicality: While theoretical formulas are provided, actual numerical computation remains difficult

Impact

1. Theoretical Contribution: Fills an important gap in the arithmetic theory of $X_1(N)$ modular curves
2. Methodological Value: Developed techniques may be applicable to other arithmetic geometry problems
3. Future Research: Provides important tools and insights for further research in related fields

Applicable Scenarios

1. Theoretical Research: Research in modular forms, arithmetic geometry, and arithmetic geometry
2. Height Theory: Study of Faltings heights and related arithmetic invariants
3. Diophantine Geometry: Research on Bogomolov-type conjectures and rational point distribution problems

References

The paper cites 46 important references, primarily including:

1. Arakelov, S.J. (1974) - Foundational work in Arakelov theory
2. Abbes, A. & Ullmo, E. (1997) - Pioneering research on the $X_0(N)$ case
3. Faltings, G. (1984) - Foundations of arithmetic surface theory
4. Iwaniec, H. (2002) - Standard reference for spectral methods in automorphic forms
5. Jorgenson, J. & Kramer, J. series - Modern developments in Green function theory

This paper represents high-level research in arithmetic geometry, solving an important theoretical problem through deep technical innovation and making significant contributions to the development of related fields.