On the second sectional \(H\)-arithmetic genus of polarized manifolds (Q2484058)

Let \(X\) be a projective variety of dimension \(n\) over the complex number field, and let \(L\) be an ample line bundle on \(X\). Then one says that the pair \((X,L)\) is a polarized variety. Many authors [e.g. \textit{T. Fujita}, ``Classification theories of polarized varieties'', Lond. Math. Soc. Lect. Note Series 155 (1990; Zbl 0743.14004)] studied polarized varieties using some numerical invariants such as the degree \(L^n\), the sectional genus \(g(L)\) and the \(\Delta\)-genus \(\Delta(L)\). In the previous paper [Commun. Algebra 32, 1069--1100 (2004; Zbl 1068.14008)], the author introduced, as a new invariant, the \(i\)-th sectional geometric genus \(g_i(X,L)\) of \((X,L)\). When \(X\) is smooth and the linear system \(| L| \) is base point free, then \(g_i(X,L)\) is the geometric genus of the transversal intersection \(X_{n-i}\) of general \(n-i\) members of \(| L| \). This accounts for the reason and the interest in this invariant. Similarly, in the paper under review, the author introduces another new invariant, i.e. the \(i\)-th sectional \(H\)-arithmetic genus \(\mathcal \chi_i^H(X,L)\) of \((X,L)\), for \(0\leq i\leq n\). To this aim one first defines the numbers \(\chi_j(X,L)\) by the identity \(\chi(tL)=\sum_{j=0}^{n}\chi_j(X,L)t^{[j]}/j!\), where \(\chi(tL)\) denotes the Euler-Poincaré characteristic of \(tL\), \(t^{[0]}=1\), and \(t^{[j]}=t(t+1)\dots(t+j-1)\) for \(j\geq 1\). Next one defines \( \mathcal \chi_i^H(X,L)=\chi_{n-i}(X,L). \) When \(X\) is smooth and the linear system \(| L| \) is base point free, then one has \(\mathcal \chi_i^H(X,L)=\chi(\mathcal O_{X_{n-i}})\), i.e. this invariant reduces to the usual arithmetic genus [in the sense of \textit{F. Hirzebruch}, ``Topological methods in algebraic geometry'', Springer-Verlag (1966; Zbl 0138.42001), 15.5, Section 15, Chapter IV]. Therefore one expects that \(\mathcal \chi_i^H(X,L)\) has analogous properties of the arithmetic genus of \(i\)-dimensional varieties. Inspired by the theory of surfaces (e.g. by Castelnuovo-De Franchis-Enriques and Bogomolov-Miyaoka-Yau inequalities), the author analyzes the case \(i=2\) and proposes the following conjectures concerning a smooth polarized variety \((X,L)\) of dimension \(n\geq 3\): (1) \(\mathcal \chi_2^H(X,L)>0\) if \(\kappa(K_X+(n-2)L)\geq 2\); (2) \(\mathcal \chi_2^H(X,L)\geq 0\) if \(0\leq \kappa(K_X+(n-2)L)\leq 1\); (3) \(\mathcal \chi_2^H(X,L)=1-q(X)\) if \(\kappa(K_X+(n-2)L)=-\infty\) and \((X,L)\) is not a scroll over a smooth surface; (4) \(8\mathcal \chi_2^H(X,L)\geq (K_X+(n-2)L)^2L^{n-2}\) if \(\kappa(K_X+(n-2)L)=-\infty\) and \((X,L)\) is neither \((\mathbb{P}^n, \mathcal O_{\mathbb{P}^n}(1))\) nor a scroll over a smooth surface; (5) \(9\mathcal \chi_2^H(X,L)\geq (K_X+(n-2)L)^2L^{n-2}\) if \(\kappa(K_X+(n-2)L)\geq 0\) (here \(\kappa(K_X+(n-2)L)\) denotes the Iitaka dimension of \(K_X+(n-2)L\)). Next the author proves (2), (3) and (4), and gives partial results about (1) and (5).