Effective bounds for very ample line bundles (Q1911141)

Let \(L\) be an ample line bundle on a nonsingular projective \(n\)-fold \(X\). A well-known conjecture of T. Fujita asserts that \(K_X+ (n+1)L\) is generated by global sections and \(K_X+ (n+2)L\) is very ample. For \(n=2\) this follows from I. Reider's theorem and the global generation part of the conjecture was proved for \(n=3\) by \textit{L. Ein} and \textit{R. Lazarsfeld} [J. Am. Math. Soc. 6, No. 4, 875-903 (1993; Zbl 0803.14004)]. The present paper is mainly concerned with the very ampleness part of the conjecture. In a previous paper [J. Differ. Geom. 37, No. 2, 323-374 (1993; Zbl 0783.32013)] the author proved that \(2K_X+12n^nL\) is very ample, using an analytic method based on the solution of a Monge-Ampère equation. In the present paper, improving a method of \textit{Y.-T. Siu} [Invent. Math. 124, No. 1-3, 563-571 (1996; Zbl 0853.32034)] based on a combination of the Riemann-Roch formula with the vanishing theorem of \textit{A. M. Nadel} [Ann. Math., II. Ser. 132, No. 3, 549-596 (1990; Zbl 0731.53063)] the author proves that \(2K_X+ mL\) is very ample for \(m\geq 2+ {3n+1 \choose n}\) and that \(m(K_X+ (n+2)L)\) is very ample for \(m\geq {3n+1 \choose n}-2n\). The method of proof gives, as a byproduct, the well-known fact that \(K_X+ (n+1)L\) is numerically effective (a result originally proved as a consequence of Mori theory). The paper also contains a refinement of a method developed by \textit{Y.-T. Siu} [Ann. Inst. Fourier 43, No. 5, 1387-1405 (1993; Zbl 0803.32017)] which enables the author to obtain a better effective Matsusaka big theorem.