Residue currents with prescribed annihilator ideals on singular varieties (Q2512970)

Let \(f=(f_1, ..., f_q)\) be a tuple of holomorphic functions defining a complete intersection in a complex manifold. Then the duality theorem for Coleff-Herrera products states that the annihilator of the Coleff-Herrera product \[ \mu^f:= \overline{\partial} \frac{1}{f_q} \wedge \ldots\wedge \overline{\partial} \frac{1}{f_1} \] equals locally the ideal generated by \(f\), i.e., \[ \text{ann}_\mathcal{O}\;\mu^f = \mathcal{J}(f_1, ..., f_q). \] In an important paper, \textit{M. Andersson} and \textit{E. Wulcan} [Ann. Sci. Éc. Norm. Supér. (4) 40, No. 6, 985--1007 (2007; Zbl 1143.32003)] generalized the construction of the Coleff-Herrera product from complete intersection ideals to arbitrary ideals of holomorphic functions. From a hermitian resolution of an ideal \(\mathcal{J}\), they constructed explicitly a vector-valued current \(R^\mathcal{J}\) such that \[ \text{ann}_\mathcal{O} \;R^\mathcal{J}= \mathcal{J}. \] If \(\mathcal{J}=\mathcal{J}(f_1, ..., f_q)\) is a complete intersection, then the current \(R^\mathcal{J}\) equals the Coleff-Herrera product. The importance of the currents of Andersson and Wulcan, which we may call Andersson-Wulcan currents, lies in the following fact. They allow the generalization of the numerous applications of Coleff-Herrera products -- particularly in division, interpolation and \(\overline{\partial}\)-homotopy formulas -- to ideals which are not complete intersections. See [\textit{M. Andersson} and \textit{H. Samuelsson}, Invent. Math. 190, No. 2, 261--297 (2012; Zbl 1271.32009); \textit{M. Andersson} and \textit{H. Samuelsson}, J. Funct. Anal. 261, No. 3, 777--802 (2011; Zbl 1227.32010); \textit{M. Andersson} et al., Ann. Inst. Fourier 60, No. 2, 417--432 (2010; Zbl 1200.32007); Zbl 1143.32003; \textit{M. Andersson} and \textit{E. Wulcan}, Invent. Math. 200, No. 2, 607--651 (2015; Zbl 1333.32006); \textit{J. Sznajdman}, ``A Briancon-Skoda Type Result for a Non-Reduced Analytic Space'', Preprint, Göteborg, \url{arXiv:1001.0322}]. In the paper under review, the author generalizes the construction from [Zbl 1143.32003] to obtain also annihilator-currents for ideals of holomorphic functions on singular varieties (Theorem 3.2 and Theorem 5.3). More precisely, let \(Z\) be an analytic subvariety of an open set in \(\mathbb{C}^n\) and \(\mathcal{J}\) an ideal of holomorphic functions in \(\mathcal{O}_Z\). Then Lärkäng constructs a current \(R^\mathcal{J}_Z\) of Andersson-Wulcan type on the variety \(Z\) such that \[ \text{ann}_{\mathcal{O}_Z}\;R^\mathcal{J}_Z = \mathcal{J}. \] It is an important observation that the construction is universal in the following sense: if \(Z\) is regular, i.e., a complex submanifold, then this current \(R^\mathcal{J}_Z\) equals the Andersson-Wulcan current defined intrinsically on the complex manifold \(Z\) modulo the \textit{structure form} (see Proposition 4.1). We should also mention an interesting development that is somewhat hidden in the paper. In [Zbl 1271.32009], Andersson and Samuelsson introduced the important concept of a \textit{structure form} that generalizes the Poincaré residue. These structure forms play a crucial role in the understanding of holomorphic forms of top-degree on singular varieties (and holomorphic forms of top-degree are of utmost importance in Complex Geometry). However, the definition (respectively construction) of structure forms of Andersson and Samuelsson is carried out in [Zbl 1271.32009] only for varieties of pure dimension. But, in order to construct the current \(R^\mathcal{J}_Z\) also for varieties \(Z\) which are not of pure dimension (see Theorem 5.3), structure forms are needed also for such varieties. So, in the present paper, Lärkäng generalizes the notion of structure forms also to varieties which are not of pure dimension (see Proposition 5.2). This is not trivial and leads to a surprisingly difficult concept (see Example 5).