Dimensional reduction and the equivariant Chern character (Q2414161)

In order to understand elliptic cohomology in geometric terms Stolz and Teichner developed the framework of supersymmetric Euclidean field theories in [\textit{S. Stolz} and \textit{P. Teichner}, Lond. Math. Soc. Lect. Note Ser. 308, 247--343 (2004; Zbl 1107.55004)]. Euclidean field theories in this context are functors from certain cobordism categories of supermanifolds to the category of topological vector spaces. More specifically the objects in the entering cobordism category of a Euclidean field theory are supermanifolds \(M\) of a specific dimension \((d|\delta)\) with a piecewise smooth reference map from the smooth underlying manifold \(|M|\) to an a priori fixed smooth manifold \(X\). The pair \((d, \delta)\) then is also called the dimension of the Euclidean field theory. Dimensional reduction as mentioned in the title of this article corresponds to constructing a \((0|1)\)-dimensional Euclidean field theory from a \((1|1)\)-dimensional Euclidean field theory by a procedure which roughly comes from taking a \((0|1)\)-dimensional supermanifold with a reference map to \(X\) and evaluate a given \((1|1)\)-dimensional Euclidean field theory on a corresponding object which essentially is the product of the given object with the circle \(S^1\), but one needs to deal with the technical issue that the product of a \((0|1)\)-dimensional supermanifold and the circle \(S^1\) is not a \((1|1)\)-dimensional supermanifold in a canonical way. In the paper under review the author introduces a version of the dimensional reduction construction where the base manifold \(X\) can be replaced by an orbifold \(\mathfrak{X}\) or, more generally, by any stack on the site of supermanifolds. His construction is based on a series of functors between variants of the Euclidean bordism categories over \(\mathfrak{X}\) which involve the inertia stack \(\Lambda \mathfrak{X}\). Formally he produces a map from the space of \((1|1)\)-dimensional Euclidean field theories over \(\mathfrak{X}\) to the space of \((0|1)\)-dimensional field theories over \(\Lambda \mathfrak{X}\). The construction itself is technically involved, so we do not give details here. In particular it takes care of the technical difficulty mentioned above, but more importantly, it also meets a design criterion, namely that it models a generalization of the equivariant Chern character. To be more specific: If \(\mathfrak{X} = X//G\) is the global quotient of a smooth action of a finite group \(G\) on a smooth manifold \(X\) then the space of \((1|1)\)-dimensional Euclidean field theories over \(\mathfrak{X}\) models the \(G\)-equivariant \(K\)-theory group \(K_G^*(X)\), while the space of \((0|1)\)-dimensional Euclidean field theories over \(\Lambda\mathfrak{X}\) models the even part of the delocalized equivariant (de Rham) cohomology \(H^{ev}_{G}(\hat{X}; \mathbb{C})\) with \(\hat{X}=\{(x,g) \in X\times G: gx=x\}\). And in view of these translations the dimensional reduction map of the paper under review models the delocalized Chern character introduced in [\textit{P. Baum} and \textit{A. Connes}, Pap. Dedic. Itiro Tamura, 163--232 (1988; Zbl 0656.55005)]. \textit{D. Berwick-Evans} [``Twisted equivariant differential K-theory from gauged supersymmetric mechanics'', Preprint, \url{arXiv:1510.07893}] produced a somewhat related version of the dimensional reduction construction. Its definition is inspired by ideas coming from perturbative quantum field theory and allows for certain twistings. However, the more geometric approach to the dimensional reduction map presented in the paper under review has already been extended to twisted settings in [\textit{A. Stoffel}, Commun. Math. Phys. 367, No. 2, 417--453 (2019; Zbl 1414.81162)].