Multiplicities of the Betti map associated to a section of an elliptic surface from a differential-geometric perspective

Abstract: For the study of the Mordell-Weil group of an elliptic curve

over a complex function field of a projective curve

B

, the first author introduced the use of differential-geometric methods arising from K"ahler metrics on

m a t h c a l H i m e s m a t h b b C

invariant under the action of the semi-direct product

m S L (2, m a t h b b R) l t i m e s m a t h b b R^{2}

. To a properly chosen geometric model

p i : m a t h c a l E o B

of

as an elliptic surface and a non-torsion holomorphic section

s i g m a : B o m a t h c a l E

there is an associated ``verticality $e t a_{s} i g m a$ of $s i g m a$ related to the locally defined Betti map. The first-order linear differential equation satisfied by $e t a_{s} i g m a$ , expressed in terms of invariant metrics, is made use of to count the zeros of $e t a_{s} i g m a$ , in the case when the regular locus $B^{0} s u b s e t B$ of $p i : m a t h c a l E o B$ admits a classifying map $f_{0}$ into a modular curve for elliptic curves with level- $k$ structure, $k g e 3$ , explicitly and linearly in terms of the degree of the ramification divisor $R_{f_{0}}$ of the classifying map, and the degree of the log-canonical line bundle of $B^{0}$ in $B$ . Our method highlights $m d e g (R_{f_{0}})$ in the estimates, and recovers the effective estimate obtained by a different method of Ulmer-Urz'ua on the multiplicities of the Betti map associated to a non-torsion section, noting that the finiteness of zeros of $e t a_{s} i g m a$ was due to Corvaja-Demeio-Masser-Zannier. The role of $R_{f_{0}}$ is natural in the subject given that in the case of an elliptic modular surface there is no non-torsion section by a theorem of Shioda, for which a differential-geometric proof had been given by the first author. Our approach sheds light on the study of non-torsion sections of certain abelian schemes.

Review text

The paper under review concerns specific complex elliptic surfaces $\mathcal E$ and their sections, namely those arising from an elliptic modular surface (as introduced by \textit{T. Shioda} [J. Math. Soc. Japan 24, 20--59 (1972; Zbl 0226.14013)]) by a base change (phrased in terms of classifying maps of the underlying modular curve). Notably, elliptic modular surfaces always have finite group of sections (and maximal Picard number, equalling Lefschetz' upper bound $h^{1,1}$). Hence the authors concentrate on non-torsion sections $\sigma$ of $\mathcal E$. In [Proc. Symp. Pure Math. None, 335--396 (1991; Zbl 0736.53048)], the first author introduced differential geometric methods to approach this problem using the ramification divisor of the classifying map. Further input comes from [\textit{N. Mok} and \textit{W.-K. To}, J. Reine Angew. Math. 444, 29--78 (1993; Zbl 0781.14030)] where first order differential equations are studied for what the authors call the verticality $\eta_\sigma$ of the section $\sigma$. Presently the authors focus on the places where the Betti multiplicities of $\sigma$ are greater than one. Their finiteness has been proven in [\textit{P. Corvaja} et al., J. Reine Angew. Math. 782, 1--41 (2022; Zbl 1485.14085)]. Here the authors develop an equality for their sum in terms of the ramification indices of the classifying map and a particular Kähler form on $\mathcal E$. Independently, such an estimate has been obtained in greater generality, and by different methods, in [\textit{D. Ulmer} and \textit{G. Urzúa}, Int. Math. Res. Not. 2021, No. 6, 4768--4802 (2021; Zbl 1477.14058)].

Recommendations

Cites work

Cited in

(1)

Betti numbers of real elliptic surfaces

This page was built for publication: Multiplicities of the Betti map associated to a section of an elliptic surface from a differential-geometric perspective

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6157823)