1
$\begingroup$

Let $X$ be a differentiable stacks, and let $(G_{0}, G_{1}, s,t)$ be a Lie groupoid representing $X$. Let $NG_{\bullet}$ be the nerve of the above groupoid. The De rham complex of $X$ can be defined as the total complex of the cosimplicial dg algebra $A_{DR}^{*}(NG_{\bullet})$ (see https://www.math.ubc.ca/~behrend/CohSta-1.pdf), where $A_{DR}(-)$ is the ordinary De Rham functor on smooth manifolds. I wonder if there is a notion of integration along the fiber (of representable morphisms)/pusforward in these settings. More concretely: for a smooth fiber bundle map $p\: : \: E\to M$ with dimension $k$ (assuming that the fiber has no boundary) we may define a map $\int_{p}\: : \: A_{DR}(E)\to A_{DR}(M)$ of degree $k$ given by "vertical" integration. This map induces the push forward $p_{*}$ in cohomology. Is there a way to define the integration along the fiber between Lie groupoids? References?

P.S: A definition of vector bundle on topological stack may be found at http://arxiv.org/abs/0712.3857.

$\endgroup$
3
  • $\begingroup$ Yes, see Alan Weinstein's “The volume of a differentiable stack”. $\endgroup$ Commented Mar 22, 2016 at 10:44
  • $\begingroup$ I check, but, what will be the integration along the fiber formula? $\endgroup$
    – Cepu
    Commented Mar 22, 2016 at 14:25
  • $\begingroup$ If you mean how the pushforward map is defined, then since you're talking about representable morphisms in your post, there is even no need to invoke Weinstein's ideas (which are need only when you have nonrepresentable fibers). In the language of functor of points, one must define the pushforward map for any S-point S→M. The pullback of E→M to S→M is a smooth submersion T→S (with oriented fibers, say), and we can integrate over its fibers as usual. This construction is functorial with respect to S and therefore defines a pushforward of forms on stacks. $\endgroup$ Commented Mar 23, 2016 at 7:17

0

You must log in to answer this question.

Browse other questions tagged .