Numerical adjunction formulas for weighted projective planes and lattice point counting (Q318159): Difference between revisions

In this extremely interesting article the authors study the general problem of counting lattice points in a polyhedral with rational vertices, and its connection with the singularity theory of surfaces and (numerical) adjunction formulas for curves in the weighted projective plane. Let us recall some notions before we present the main results. For a polygon \(\mathcal{P}\) or its dilations \(d\mathcal{P} = \{dp: p\in \mathcal{P}\}\) the number of integral points (resp. rational points) is a polynomial (resp. quasi-polynomial) in \(d\) of degree \(\dim \mathcal{P}\) referred to as the Ehrhart (quasi)-polynomial. Let \[ dD_{w} = D_{w,d} = \big\{(x,y,z) \in \mathbb{R}^{3} : x,y,z \geq 0, \, w_{0}x + w_{1}y + w_{2}z =d \big\}, \] where \(w = (w_{0},w_{1},w_{2})\) are pairwise coprime and \(D_{w,1}\) is a rational polygon. Moreover, for a given \(w\) one defines \(|w| = w_{0} + w_{1} + w_{2}\) and \(\bar{w} = w_{0}w_{1}w_{2}\). The key ingredient to connect rational polygons with some arithmetic problems comes from the observation that \[ L_{w}(D):= \#(D_{w,d} \cap \mathbb{Z}^{3}) = h^{0}(\mathbb{P}^{2}_{w}, \mathcal{O}(d)), \] that is, the dimension of the vector space of weighted homogeneous polynomials of degree \(d\). One of the main results is devoted to a numerical adjunction formula. Assume that \(C \subset \mathbb{P}^{2}_{w}\) is a quasismooth curve of degree \(d\). Using the classical adjunction formula and the weighted Bézout theorem one has \[ 2g(C) -2 = \frac{d(d-|w|)}{\bar{w}}. \] It can be shown that the generic curve of degree \(k\bar{w}\) is smooth and in this case \[ (\star): \qquad h^{0}(\mathbb{P}^{2}_{w},\mathcal{O}(k\bar{w}-|w|)) = L_{w}(k\bar{w}-|w|) = g_{w,k\bar{w}}, \] where \(g_{w,t}:= \frac{t(t-|w|)}{2\bar{w}} + 1\). However, for a general \(d\), the generic curve of degree \(d\) in \(\mathbb{P}^{2}_{w}\) is not necessarily quasismooth. The main goal of the authors is to revisit \((\star)\) in the singular case. The main result of the paper can be formulated as follows. Theorem 1. The Ehrhart quasipolynomial \(L_{w}(d)\) for \(D_{w,1}\) satisfies \[ L_{w}(d) = g_{w,d+|w|} - \sum_{P \in \text{Sing}(\mathbb{P}^{2}_{w})} \triangle_{P}(d+|w|). \] The number \(g_{w,k} = \frac{k(k-|w|)}{2\bar{w}}+1\) is called the virtual genus and \(\triangle_{P}(k)\) is a periodic function being an invariant associated to the singularity \(P \in \mathbb{P}^{2}_{w}\) -- the notion is rather technical so we refer to Definition \(3.8\) for a precise formulation. It turns out that the number \(\triangle_{P}(k)\) has a geometrical interpretation and it can be computed using invariants of curve singularities on a singular surface. The first invariant \(\delta_{P}(f)\) for a given curve \((\{f = 0\},P)\) can be computed using the number of local branches of a curve \(C = \{f=0\}\) and the orbifold Euler characteristic of the Milnor fiber (Definitions \(3.1\) and \(3.2\)), and the second (new) invariant \(\kappa_{P}(f)\) is the dimension of a certain quotient of modules (i.e., the module of \(k\)-invariant germs \(\mathcal{O}_{X,P}(k)\) by a certain logarithmic (sub)module -- see Definitions \(2.1\), \(2.3\), \(3.4\) and \(3.5\)). Theorem 2. Let \((f,P)\) be a reduced curve germ at \(P \in X\), a surface cyclic quotient singularity. Then \[ \triangle_{P}(k) = \delta_{P}(f) - \kappa_{P}(f) \] for any reduced germ \(f \in \mathcal{O}_{X,P}(k).\) Finally, using the above results one can show the following numerical adjunction formula for a general singular curve \(C \subset \mathbb{P}^{2}_{w}\), its genus \(g(C)\), and \(\kappa_{P}\). Theorem 3. Let \(C = \{f=0\} \subset \mathbb{P}^{2}_{w}\) be an irreducible curve of degree \(d\). Then \[ h^{0}(\mathbb{P}^{2}_{w}, \mathcal{O}(d-|w|)) = g(C) + \sum_{P \in \text{Sing}(C)}\kappa_{P}(f). \] The paper is really nicely written (even taking into account that there are some technical parts and technical definitions) and contains detailed examples. It is worth pointing out that a very nice introduction to the subject can be found in the PhD thesis of the third author [Algebraic and Topological Invariants of Curves and Surfaces with Quotient Singularities. Universidad de Zaragoza (PhD thesis) (2013), \url{https://zaguan.unizar.es/record/11738}]. The reviewer strongly encourages interested readers to consult details and other results there.