Cookies

We use cookies to ensure that we give you the best experience on our website. You can change your cookie settings at any time. Otherwise, we'll assume you're OK to continue.

Department of Mathematical Sciences

Seminar Archives

On this page you can find information about seminars in this and previous academic years, where available on the database.

Pure Maths Colloquium: Hodge type theorems for arithmetic manifolds associated to orthogonal and unitary groups.

Presented by John Millson, Maryland

17 August 2015 16:00 in PH30 (Physics)

I will speak about two papers with Nicolas Bergeron and Colette Moeglin (and one paper with Bergeron, Zhiyuan Li and Moeglin). For each n between 1 and p, the (standard) arithmetic manifolds M associated to the orthogonal groups SO(p,q) resp. the unitary groups SU(p,q) contain many totally geodesic submanifolds N of codimension nq associated to embedded subgroups SO(p-n,q) resp SU(p-n,q). In the 1980's Steve Kudla and I proved that the cohomology classes dual to such N could be represented by differential forms constructed using the Weil or oscillator representation. More precisely, for the case of SO(p,q), the theory of the oscillator representation provides an integral transform ( "geometric theta lifting") from the space of holomorphic Siegel modular forms for Sp(2n,R) and weight (p+q)/2 to harmonic differential nq-forms on the above manifolds M. Recently using Arthur's work on the Selberg trace formula we proved that the geometric theta lifting was onto. In my talk I will explain the basic differential-geometric principles behind the geometric theta lifting.

The theory of the previous paragraph has the following applications:

1.For the case of SO(p,1,) p >3,, the next-to-top homology group H_{p-1}(M) for the standard arithmetic real hyperbolic p-manifolds M is spanned by totally-geodesic hypersurfaces.

2. For the case of SU(p,1), the Hodge and Tate conjectures hold away from the middle third of cohomological degrees.

3. For the case of SO(2,19), the Noether-Lefschetz conjecture of Maulik and Pandharapande holds (Noether-Lefschetz divisors span the Picard group).

Contact anna.felikson@durham.ac.uk for more information