Ultrametric theta functions and abelian varieties (Q4133743)

From MaRDI portal

!

This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:

scientific article; zbMATH DE number 3557937
Language Label Description Also known as
default for all languages
No label defined
    English
    Ultrametric theta functions and abelian varieties
    scientific article; zbMATH DE number 3557937

      Statements

      Ultrametric theta functions and abelian varieties (English)
      0 references
      1978
      0 references
      Let \(k\) be a field complete with respect to a non-archimedean valuation. Let \(g\) be a positive integer and \((a_{ij})\) be a \(g\times g\) matrix with entrie in \(k^*\) satisfying the following ``Riemann conditions'': \(a_{ij}=a_{ji}\) for all \(i, j\) and the matrix (ord\(a_{ij}\)) is positive definite. Following McCabe [``P-adic theta functions''. Ph. D. thesis, Harvard University (1968)] a graded ring \(R\) of theta functions associated to \((a_{ij})\) is constructed, and it ja shown that \(A=\)Proj\((R)\) is an abelian variety of dimension \(g\). If \(\Gamma\) is the subgroup of \(G_g = (k^*)^g\) generated by the column vectors of \((a_{ij})\), a homomorphism \(\varphi:G_g/\Gamma\to A_k\) is defined. The main purpose of the paper is to show that \(\varphi\) is bijective using only simple calculations with Laurent power series. In order to prove that \(\varphi\) is bijective, we first treat the ``diagonal case'', i.e. the case where the off-diagonal elements of \((a_{ij})\) have order 0. The diagonal case is the heart of the paper. Here the reduction \(\bar A\) of \(A\) plays a central role. \(\bar A\) looks very much like the product of rational curves with an ordinary double point. To each \(P\in A_k\), we associate a subset \(S(P)\) of \(\{1,2,\dots,g\}\) which describes how singular \(P\) is on \(\bar A\). An implicit function type argument is used to show that \(\varphi\) is injective and that the points with \(S(P)=\emptyset\) are in \(\varphi(G_g)\). The surjectivity of \(\varphi\) is proved by induction on the cardinality of \(S(P)\). The main ingredients are: an addition formula on \(A\) and the Decomposition Theorem (theorem II. 6.6) for the points of \(A\). For a more general situation, the diagonal case and an isogeny argument are used to show that \(\varphi\) is bijective assuming only that each ord\((a_{ij})\) is rational. This mild restriction is unnecessary as Gerritzents results show [\textit{L. Gerritzen}, Math. Ann. 196, 323--346 (1973; Zbl 0255.14013)], but we have been unable to avoid it. The question considered above was first treated by \textit{H. Morikawa} [Nagoya Math. J. 20, 1--27 (1962; Zbl 0115.39001)]. J. Tate gave a complete solution for the case \(g= 1\). McCabe (loc. cit.) proved that \(\varphi\) is injective and it is surjective if \(k\) is locally compact. \textit{D. Mumford} [Compos. Math. 24, 239--272 (1972; Zbl 0241.14020)], \textit{M.Raynaud} [Actes Congr. internat. Math. 1970, 1, 473--477 (1971; Zbl 0223.14021)] and Gerritzen (loc. cit.) have generalized and solved the problem using non-elementary techniques.
      0 references

      Identifiers