Abelian branched covers of rational surfaces (Q6176265)

In the paper under review the authors study abelian covers of rational surfaces branched along line arrangements, mostly towards the geography problem for closed simply connected non-spin irreducible symplectic \(4\)-manifolds with positive signature. Let \(M\) be a compact smooth \(4\)-dimensional manifold without boundary. Let \(e(M)\) denotes the Euler characteristic and \(\sigma(M)\) the signature of \(M\). One defines \(\chi_{h}(M)=\frac{1}{4}(e(M)+\sigma(M))\) and \(c_{1}^{2}(M) = 2e(M)+3\sigma(M)\). When \(M\) is a complex surface, \(\chi_{h}\) is the holomorphic Euler characteristic of \(M\) and \(c_{1}^{2}(M)\) is the square of the first Chern class of \(M\). We say that \(M\) is irreducible if every connected sum decomposition \(M = X \# Y\) implies that either \(X\) or \(Y\) is homeomorphic to the \(4\)-sphere \(S^4\). The authors study the geography problem. Given a pair of integers \((a,b)\) satisfying \(a\geq 1\) and \(b\geq 0\), we ask whether there exists a compact simply connected non-spin irreducible symplectic \(4\)-manifold \(M\) without boundary satisfying \(\chi_{h}(M) = a\) and \(c_{1}^{2}(M)=b\). We note that \(M\) must satisfy \[b=8a + \sigma(M), \quad a\geq \frac{1}{2}(\sigma(M)+1).\] Here the authors assume that \(b-8a\geq 0\), so the signature is non-negative. To stay the main result of the paper, let us recall two definitions. Definition 1. We say that a \(4\)-manifold \(M\) has the \(\infty^{2}\)-property if there exists infinitely many pair-wise non-diffeomorphic irreducible symplectic \(4\)-manifolds and infinitely many pair-wise non-diffeomorphic irreducible non-symplectic \(4\)-manifolds, all of which are homeomorphic to \(M\). Definition 2. Given an integer \(\sigma \geq 0\), let \(\lambda(\sigma)\) be the smallest positive integer with the following two properties: i) \(\lambda(\sigma) \geq \lceil (\sigma + 1)/2\rceil,\) ii) Every integral point \((a,b)\) on the line \(b=8a+\sigma\) satisfying \(a\geq \lambda(\sigma)\) is realized as \((\chi_{h}(M), c_{1}^{2}(M))\), where \(M\) is a compact simply connected non-spin \(4\)-manifold without boundary that has the \(\infty^{2}\)-property. The main result of the paper under review provides the best known asymptotic upper-bound for \(\lambda(\sigma)\), namely the authors show that \[\limsup_{\sigma \rightarrow \infty} \frac{\lambda(\sigma)}{\sigma} \leq \frac{17}{11}.\] In order to obtain the above upper-bound, the authors construct infinitely many new symplectic \(4\)-manifolds with positive signature that have the \(\infty^2\)-property. The new building blocks are complex surfaces that are abelian covers of rational surfaces branched along complex line arrangements.