skip to content
 

Ricci-flat manifolds and a spinorial flow

Presented by: 
Bernd Ammann Universität Regensburg
Date: 
Monday 27th June 2016 - 10:00 to 11:00
Venue: 
INI Seminar Room 1
Abstract: 
Joint work with Klaus Kröncke, Hartmut Weiß and Frederik Witt

We study the set of all Ricci-flat Riemannian metrics on a given compact manifold M.
We say that a Ricci-flat metric on M is structured if its pullback to the universal cover admits a parallel spinor. The holonomy of these metrics is special as these manifolds carry some additional structure, e.g. a Calabi-Yau structure or a G<sub>2</sub>-structure.

The set of unstructured Ricci-flat metrics is poorly understood.  Nobody knows whether unstructured compact Ricci-flat Riemannian manifolds exist, and if they exist, there is no reason to expect that the set of such metrics on a fixed compact manifold should have the structure of a smooth manifold.

On the other hand, the set of structured Ricci-flat metrics on compact manifolds is now well-understood.

The set of structured Ricci-flat metrics is an open and closed subset in the space of all Ricci-flat metrics.
The holonomy group is constant along connected components.
The dimension of the space of parallel spinors as well.
The structured Ricci-flat metrics form a smooth Banach submanifold in the space of all metrics.
Furthermore the associated premoduli space is a finite-dimensional smooth manifold.

These results build on previous work by J. Nordström, Goto, Koiso, Tian & Todorov, Joyce, McKenzie Wang and many others.
The important step is to pass from irreducible to reducible holonomy groups.

In the last part of the talk we summarize work on the L<sup>2</sup>-gradient flow of the functional $(g,\phi)\mapsto E(g,\phi):=\int_M|\nabla^g\phi|^2$.
This is a weakly parabolic flow on the space of metrics and spinors of constant unit length. The flow is supposed to flow against structured Ricci-flat metics. Its geometric interpretation in dimension 2 is some kind of Willmore flow, and in dimension 3 it is a frame flow.
We find that the functional E is a Morse-Bott functional. This fact is related to stability questions.

Associated publications:
http://www.mathematik.uni-regensburg.de/ammann/preprints/holrig
http://www.mathematik.uni-regensburg.de/ammann/preprints/spinorflowI
http://www.mathematik.uni-regensburg.de/ammann/preprints/spinorflowII  
The video for this talk should appear here if JavaScript is enabled.
If it doesn't, something may have gone wrong with our embedded player.
We'll get it fixed as soon as possible.
University of Cambridge Research Councils UK
    Clay Mathematics Institute London Mathematical Society NM Rothschild and Sons