Limit multiplicities for principal congruence subgroups of GL(n) and SL(n)

Abstract

We study the limiting behavior of the discrete spectra associated to the principal congruence subgroups of a reductive group over a number field. While this problem is well understood in the cocompact case (i.e., when the group is anisotropic modulo the center), we treat groups of unbounded rank. For the groups GL(n) and SL(n) we show that the suitably normalized spectra converge to the Plancherel measure (the limit multiplicity property). For general reductive groups we obtain a substantial reduction of the problem. Our main tool is the recent refinement of the spectral side of Arthur's trace formula obtained in [FLM11, FL11], which allows us to show that for GL(n) and SL(n) the contribution of the continuous spectrum is negligible in the limit.

49 pages slightly enhanced version fixed several inaccuracies in first version

References in corpus (1)

• Dual R-groups of the inner forms of SL(N)

Cited by in corpus (9)

• An approximation principle for congruence subgroups
• Rank, combinatorial cost and homology torsion growth in higher rank lattices
• Sato-Tate Equidistribution for Families of Automorphic Representations through the Stable Trace Formula
• Gamma factors for the Asai cube representation
• Strong Limit Multiplicity for arithmetic hyperbolic surfaces and $3$-manifolds
• An approximation principle for congruence subgroups II: application to the limit multiplicity problem
• The density hypothesis for horizontal families of lattices
• On the analytic properties of intertwining operators II: local degree bounds and limit multiplicities
• On the local $L^2$-Bound of the Eisenstein Series