Divisors over determinantal rings defined by two by two minors
Abstract
Let E and G be free modules of rank e and g, respectively, over a commutative noetherian ring R. The identity map on E* tensor G induces the Koszul complex ... -> SmE* tensor SnG tensor Wedgep(E* tensor G) -> Sm+1E* tensor Sn+1G tensor Wedgep-1(E* tensor G) -> ... and its dual ... -> Dm+1E tensor Dn+1G* tensor Wedgep-1(E tensor G*) -> DmE tensor DnG* tensor Wedgep(E tensor G*)-> ... Let Hm,n,p be the homology of the top complex at Sm tensor Sn tensor Wedgep and Hm,n,p the homology of the bottom complex at Dm tensor Dn tensor Wedgep. It is known that Hm,n,p is isomorphic to Hm',n',p', provided m+m'=g-1, n+n'=e-1, p+p'=(e-1)(g-1), and m-n is between 1-e and g-1. In this paper we exhibit an explicit quasi-isomorphism M of complexes which gives rise to this isomorphism. The mapping cone of M is a split exact complex. Our complexes may be formed over the ring of integers; they can be passed to an arbitrary ring or field by base change. Knowledge of the homology of the top complex is equivalent to knowledge of the modules in the resolution of the Segre module Segre(e,g,m-n). The Segre modules are a set of representatives of the divisor class group of the determinantal ring defined by the 2 by 2 minors of an e by g matrix of indeterminants. If R is the ring of integers, then the homology Hm,n,p is not always a free abelian group. In other words, if R is a field, then the dimension of Hm,n,p depends on the characteristic of R. The module Hm,n,p is known when R is a field of characteristic zero; however, this module is not yet known over arbitrary fields.
Turn this paper into a full lesson
ArcXiv compiles a staged curriculum from this paper: 8-12 lessons across beginner → advanced, synthesised section guides, visuals, flashcards, a quiz, exercises, and on-demand deep dives per section. Grounded in the abstract, never invented.