Arbitrarily large families of spaces of the same volume (Q715199): Difference between revisions

Let \(\mathcal{G}\) be any real Lie group which either (i) has a factor not of type \(A_i, i=1,2,3\), or (ii) only has factors of type \(A_2\) or of type \(A_3\). In this paper the author constructs, for every \(m\geq 1\), a family \(\Gamma_1,\ldots,\Gamma_m\) of nonconjugated torsion-free lattices such that all quotients \(\Gamma_i\backslash \mathcal{G}\) have the same volume. This result was known for \(\mathcal{G}=\mathrm{SO}(n,1)\) [\textit{B. Zimmermann}, Monatsh. Math. 117, No. 1--2, 139--143 (1994; Zbl 0789.57008) and \textit{A. Lubotzky}, Transform. Groups 1, 71--82 (1996; Zbl 0876.22015)] and actually for all real Lie groups [\textit{D. B. McReynolds}, ``Isospectral locally symmetric manifolds'', Preprint (2009), \url{arXiv:math/060650}]. All these previous works obtained the lattices as fundamental groups of finite covers of the same degree of a given manifold, here the lattices constructed by the author do not arise in this way but they are nonetheless commensurable. His construction is quite flexible and allows to add the following features: in case (i), given an arithmetic lattice \(\Gamma\), all \(\Gamma_i\) can be chosen to be commensurable with \(\Gamma\), and in any case if \(\mathcal{G}\) contains an arithmetic non-cocompact lattice then the \(\Gamma_i\) can be chosen to be either cocompact or not. In case (ii), when \(\mathcal{G}\) has more than one factor the \(\Gamma_i\) are irreducible. The method of proof is based on the Prasad volume formula. The lattices arise as stabilizers of simplices in the Bruhat-Tits buildings at finite places; choosing different types of simplices at a finite number of places ensures that the groups are not conjugated, but the local data contributing to the formula will be the same in each case. The paper gives a brief overview of the theory leading to the volume formula but a reader not familiar with algebraic groups over local fields will have a hard time reading the proofs.