Symbolic powers of generalized star configurations of hypersurfaces (Q2062727)

Let \(R:=\mathbb{K}[x_{0},\dots, x_{n}]\) be a graded polynomial ring over an infinite field \(\mathbb{K}\) and let \(\mathfrak{m} = \langle x_{0},\dots, x_{n} \rangle\). Suppose that \(s \geq n+1\) and let \(\mathcal{F} = \{f_{1},\dots, f_{s}\}\) be a set of forms in \(R\) satisfying some (technical) genericity conditions. For a fixed choice of positive integers \(a\) and \(b\) such that \(a\leq bs\), the generalized star configuration ideal is the uniform \(a\)-fold product ideal \[I_{a}(\mathcal{F}^{b}) = I_{a}(f_{1}^{b} \cdots f_{s}^{b}).\] In general, the \(a\)-fold product ideal \(I_{a}(f_{1}^{b} \cdots f_{s}^{b})\) is generated by the \(a\)-fold products of the forms \(f_{1},\dots, f_{s}\) with multiplicities \(b_{1},\dots,b_{s}\) respectively, namely \[I_{a}(f_{1}^{b_{1}}\cdots f_{s}^{b_{s}}) := \langle f_{1}^{n_{1}} \cdots f_{s}^{n_{s}} \, : \, 0 \leq n_{1} \leq b_{i} \text{ for each } i \text{ such that } \sum_{i} n_{i} = a \rangle.\] We say that an \(a\)-fold product \(I_{a}(f_{1}^{b_{1}} \cdots f_{s}^{b_{s}})\) is uniform when the multiplicities \(b_{1},\dots, b_{s}\) are all identical, for some integer \(b\), and one writes it as \(I_{a}(\mathcal{F}^{b})\). It is treated as the generalized star configuration ideal since when \(b=1\), one will get back the star configuration ideal of hypersurfaces \[I_{c,\mathcal{F}} = \bigcap_{1 \leq i_{1} < \cdots < i_{c} \leq s} \langle f_{i_{1}},\dots, f_{i_{c}} \rangle\] for \(c = s-a+1\). Recall that for a given ideal \(I\) in a ring \(R\), the \(m\)-th symbolic powers of \(I\) is defined by \[I^{(m)} = \bigcap_{\mathfrak{p} \in\mathrm{Ass}(R/I)} I^{m}R_{\mathfrak{p}} \cap R.\] For instance, if \(I\) is the defining ideal of a reduced affine scheme over an algebraically closed field of characteristic zero, the Nagata-Zariski theorem tells us that \(I^{(m)}\) is generated by all the polynomials whose partial derivatives of orders up to \(m-1\) vanish on this scheme. Finally, for a homogeneous ideal \(J\) one defines \(\alpha(J)\) which is the least degree of non-zero forms in \(J\). The main result of the paper under review can be formulated as follows. Main Theorem. Let \(a\) and \(b\) be positive integers such that \((b-1)s +1 < a \leq bs\). Assume furthermore that \(n\geq bs - a +1\) and \(\mathcal{F} = \{f_{1},\dots, f_{s}\}\) is a set of \((bs-a+1)\)-generic \(d\)-forms in \(R\), i.e., any subset of size at most \((bs-a+2)\) will form a regular sequence. Then the following properties hold for the generalized star configuration ideal \(I = I_{a}(\mathcal{F}^{b})\) whose big height is known to be \(h=bs-a+1\): i) The symbolic powers of \(I\) are sequentially Cohen-Macaulay; ii) The ideal \(I\) satisfies the Harbourne-Huneke containment, i.e., \[I^{(\ell(h+m-1)-h+k)}\subset \mathfrak{m}^{d((\ell-1)(h-1)+k-1)(bh-(bs-a))}(I^{(m)})^{\ell}\] for all positive integers \(k,\ell, m\). iii) The inequality (being a Demailly-like bound) \[\frac{\alpha(I^{(\ell)})}{\ell} \geq \frac{\alpha(I^{(m)}) + h-1}{m+h-1}\] holds for all \(\ell, m \geq 1\).