The geodesic restriction problem for arithmetic spherical harmonics

Maximiliano Sanchez Garza · 2025

Read The geodesic restriction problem for arithmetic spherical harmonics by Maximiliano Sanchez Garza (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2509-24874.txt.

Introduction

Harmonic functions restricted to geodesics can be far smaller than general bounds allow, Maximiliano Sanchez Garza shows in a 2025 paper. The setting is the sphere, with Laplacian eigenfunctions normalized in the L2 norm. A general problem in analysis asks how the mass of such a function distributes around the manifold, and one attack studies its norms along submanifolds such as geodesics.

For a generic surface, the best general bound grows like a small power of the eigenvalue, and that bound is optimal in general. Here the eigenfunction is also an eigenfunction of all Hecke operators, an arithmetic special case. Restricting to geodesics tied to negative fundamental discriminants, the bound improves to essentially the eigenvalue raised to any positive power, which is nearly sharp.

Catalog record

CREATORMaximiliano Sanchez Garza
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE108 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2509-24874

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

The geodesic restriction problem for arithmetic spherical harmonics Maximiliano Sanchez Garza Abstract . Given a Riemannian manifold M M and an L 2 L^{2} -normalized Laplacian eigenfunction ψ \psi on M M with eigenvalue λ 2 \lambda^{2} , a general problem in analysis is to understand how the mass of ψ \psi distributes around M M . There are different ways to attack this problem. One of them is to analyze the L p L^{p} -norm of ψ \psi restricted to a submanifold of M M . Here, we concentrate on the case M = S 2 M=S^{2} , p = 2 p=2 , and we restrict to geodesics of the sphere. Burq, Gérard, and Tzvetkov showed, for γ \gamma a geodesic of S 2 S^{2} (and indeed for more general surfaces), that ‖ ψ | γ ‖ L 2 ≪ λ 1 / 4 ||\psi|_{\gamma}||_{L^{2}}\ll\lambda^{1/4} and that this bound is optimal in general. In this paper, we specialize to the case in which ψ \psi is an eigenfunction of all the Hecke operators on the sphere and consider the set of geodesics 𝒞 D \mathcal{C}_{D} of S 2 S^{2} associated to fundamental discriminants D < 0 D<0 . By combining approaches of Ali and Magee, we improve the previous upper bound to | | ψ | 𝒞 D | | L 2 ≪ D , ε λ ε ||\psi|_{\mathcal{C}_{D}}||_{L^{2}}\ll_{D,\varepsilon}\lambda^{\varepsilon} for any ε > 0 \varepsilon>0 , which is essentially sharp. 1. Introduction Let (M,g) be a Riemannian manifold and let \Delta be the Laplace operator associated to g . Let \psi be a Laplacian eigenfunction on M satisfying \Delta\psi=\lambda^{2}\psi . An interesting problem in analysis is to study how much of the mass of \psi concentrates in small subsets of M . There are several ways to approach this question; see the introduction of [ BGT07 ] for some of these approaches and references. One of them is to take \psi to be L^{2} -normalized and bound ||\psi|_{M^{\prime}}||_{L^{p}} , where M^{\prime} is a submanifold of M and 2\leq p\leq\infty . This approach was first studied by Reznikov in [ Rez04 ] in the particular case of M being a compact hyperbolic Riemann surface, p=2 , and restricting \psi to a closed geodesic \gamma , where he showed that (1) ||\psi|_{\gamma}||_{L^{2}}\ll\lambda^{\frac{1}{4}}. Later, in [ BGT07 , Thm. 3] , Burq, Gérard, and Tzvetkov generalized Reznikov’s results to Riemannian manifolds in general, where the bounds are now dependent on p and the dimensions of M and M^{\prime} , and showed that these are sharp in most cases. Sarnak, in his letter to Reznikov in [ Sar08 , p. 1] , calls bound ( 1 ) and the bounds obtained by Burq, Gérard, and Tzvetkov the convexity bounds . The reason for this is that, in the case of M^{\prime}=M being an arithmetic hyperbolic surface and p=\infty , there is a deep connection between our problem, the Ramanujan conjectures for Maass forms, and the Lindelöf Hypothesis. This connection is described in [ Sar95 , Section 4] . Even though the bound ( 1 ) is sharp in general, it is expected to be not optimal in the case where M has constant curvature -1 since the geodesic flow is ergodic and quantum ergodicity holds (see [ Shn74 ] , [ CdV85 ] , [ Zel87 ] ). The only power-saving progress in this setting is when M is an arithmetic quotient of the modular surface, such as M=\Gamma_{0}(q)\backslash\mathbb{H} , and \psi is arithmetic, i.e. a simultaneous eigenfunction of \Delta and all the Hecke operators. The first improvement to the convexity bound in this setting is due to Iwaniec and Sarnak, where they showed in [ IS95 , Thm. 0.1] that ||\psi||_{L^{\infty}}\ll_{\varepsilon}\lambda^{\frac{5}{24}+\varepsilon} for any \varepsilon>0 . More recently, Ali showed in [ AA22 ] that for any \varepsilon>0 we have (2) ||\psi|_{\mathcal{C}_{D}}||_{L^{2}}\ll_{D,\varepsilon}\lambda^{\theta+\varepsilon} where D>0 is a fundamental discriminant, \mathcal{C}_{D} is the union of all closed geodesics in \mathrm{SL}_{2}(\mathbb{Z})\backslash\mathbb{H} associated to D , and \theta\geq 0 is any bound towards the Ramanujan Conjecture for Maass forms. One can take \theta=\frac{7}{64} by results from Kim and Sarnak in [ KRS03 , Appendix 2, Prop. 2] . Figure 1. From left to right, graphic illustrations of zonal, tesseral, and sectorial spherical harmonics. If f(\xi) is the spherical harmonic in consideration, the red zones represent positive values of \mathrm{Re}(f(\xi)) , whereas the blue zones represent negative values of \mathrm{Re}(f(\xi)) . The intensity of the color represents |f(\xi)| . We concentrate on the case where M=S^{2} is the 2 -sphere, that is, S^{2}:=\{(x,y,z)\in\mathbb{R}^{3}:x^{2}+y^{2}+z^{2}=1\}, which has constant curvature 1 . We give it the round metric and associated measure, which gives it the structure of a Riemannian manifold. Let \Delta be its Laplace operator. We know that L^{2}(S^{2}) has a complete spectral resolution with respect to \Delta . Moreover, this spectral resolution is completely explicit in terms of spherical harmonics; see [ STV18 , Thm. 2.9] , [ SW71 , p. 140] , [ AH12 , Th. 2.34, Prop. 3.5] . More explicitly, the \ell -th Laplacian eigenvalue is \lambda_{\ell}:=\ell(\ell+1) and its eigenspace, denoted by H_{\ell} , is the space of harmonic homogeneous polynomials in three variables of degree \ell restricted to the sphere S^{2} , which are precisely the spherical harmonics. Let \psi\in H_{\ell} and let \gamma be a geodesic of S^{2} . We know ||\psi|_{\gamma}||_{L^{2}} satisfies bound ( 1 ). In fact, for arbitrary \psi , the bound ( 1 ) is sharp; see [ BGT07 , Thm. 1] . The reason for this is that, at least for certain \psi , there are high concentrations of mass in specific regions of the sphere (see Figure 1 ). Even though S^{2} has constant curvature 1 , we expect improvements on the bound ( 1 ) for specific \psi . The reason for this is that a basis of Laplacian eigenfunctions for S^{2} can also be diagonalized further with respect to Hecke operators, thus making the basis elements arithmetic. These operators are defined using the action of B^{\times}(\mathbb{R}) on S^{2} via rotations, where B^{\times}(\mathbb{R}) is the group of units of the Hamiltonian quaternion algebra B(\mathbb{R}) over \mathbb{R} . This action makes S^{2} into a homogeneous space. To be more precise, for any ring R , we define B(R):=\langle 1,i,j,k\>|\>i^{2}=j^{2}=k^{2}=-1,\>ij=-ji=k\rangle_{R}, so that B(R)=B(\mathbb{Q})\otimes_{\mathbb{Q}}R for any \mathbb{Q} -algebra R . We denote by B^{\times}(R) the group of units of B(R) . Then B^{\times} is an affine algebraic group over \mathbb{Q} . For \alpha=a+bi+cj+dk\in B(R) , let \mathrm{Nrd}\left(\alpha\right):=a^{2}+b^{2}+c^{2}+d^{2}\in R and \mathrm{Trd}\left(\alpha\right):=2a\in R be its (reduced) norm and (reduced) trace, respectively. Let B^{0,1}(\mathbb{R}) be the subset of B(R) of quaternions of trace zero and norm 1 . The group B^{\times}(R)\ni\alpha acts on B^{0,1}(R)\ni\gamma by conjugation, i.e. \alpha\cdot\gamma:=\alpha\gamma\alpha^{-1} . We now specialize to the case R=\mathbb{R} . Since we have the bijection 1 1 1 Throughout the paper, we identify B^{0,1}(\mathbb{R}) and S^{2} under the above bijection and use them interchangeably. \displaystyle B^{0,1}(\mathbb{R}) \displaystyle\longrightarrow S^{2} \displaystyle xi+yj+zk \displaystyle\longmapsto(x,y,z), we obtain an action of B^{\times}(\mathbb{R}) on S^{2} . This action is by rotations (see [ CS03 , §3.1] ). Hence, there is an induced action on L^{2}(S^{2}) , which is the left regular representation of B^{\times}(\mathbb{R}) in L^{2}(S^{2}) . Let \mathcal{O}\subset B(\mathbb{R}) denote the

Read the complete text