Strict comparison for \(C^*\)-algebras arising from almost finite groupoids (Q2192896)

Let \(G\) be an étale groupoid with (compact) totally disconnected unit space. For such \(G\), Matui introduced the notion of ``almost finiteness'' [\textit{H. Matui}, Abel Symp 2015, No. 12, 203--230 (2016; Zbl 1367.54021)]. Roughly speaking, this property corresponds to \(G\) being locally approximable by principal groupoids. Specializing to the case of almost finite, free, minimal group actions, Kerr was able to show that the associated crossed product \(C(X)\rtimes G\) is \(\mathcal{Z}\)-stable provided that the group is amenable and infinite [\textit{D. Kerr}, ``Dimension, comparison, and almost fniteness'', J. Eur. Math. Soc. (JEMS) 22, No.~11, 3697--3745 (2020; Zbl 1465.37010)]. Here, \(\mathcal{Z}\) denotes the so-called Jiang-Su algebra, and \(\mathcal{Z}\)-stability is conjecturally equivalent to ``strict comparison'' for separable, simple, nuclear (non-elementary) \(C^*\)-algebras (this is the Toms-Winter conjecture [\textit{W. Winter}, in: Proceedings of the international congress of mathematicians 2018, ICM 2018, Rio de Janeiro, Brazil, August 1--9, 2018. Volume III. Invited lectures. Hackensack, NJ: World Scientific; Rio de Janeiro: Sociedade Brasileira de Matemática (SBM). 1801--1823 (2018; Zbl 1462.46070)]). In the paper under review, the authors prove that, given a minimal almost finite groupoid \(G\) as in the opening, its reduced groupoid \(C^*\)-algebra has strict comparison and real rank zero. This result had been claimed (without proof) in a previous work by Suzuki who verified that almost finite minimal groupoid \(C^*\)-algebras have stable rank one [\textit{Y. Suzuki}, ``Almost finiteness for general étale groupoids and its applications to stable rank of crossed products'', Int. Math. Res. Notices 2020, No. 19, 6007--6041 (2020; \url{doi:10.1093/imrn/rny187})]. Suzuki's strategy is in essence a ``local version'' of Phillips' large subgroupoid technique [\textit{N. C. Phillips}, Commun. Math. Phys. 256, No. 1, 1--42 (2005; Zbl 1084.46056)]. The arguments in this paper heavily rely on Suzuki's main result and on Phillips' lines of proof for many intermediate steps. The strategy adopted by the authors does not work for general non-minimal almost finite groupoids as these usually do not have stable rank one. It is worth mentioning that \(C^*_r(G)\) in the main theorem of this work may have bizarre properties. Indeed, the result holds for non-separable, non-nuclear \(C^*\)-algebras (Elek constructed non-amenable, minimal, almost finite (ample) groupoids in [\textit{G. Elek}, ``Qualitative graph limit theory. Cantor dynamical systems and constant-time distributed algorithms'', Preprint (2018), \url{arXiv:1812.07511}]).