There are a fair number of papers by Beauville, Laszlo, Sorger, Kumar and others on the geometry of $LG/L^+G = G((z))/G[[z]]$ where $G$ is a simply connected and simple group over $\mathbb{C}$. In particular it is known that $Pic(LG/L^+G)$ is equal to $\mathbb{Z}$. The pull back to $LG$ of a generator $1 \in Pic(LG/L^+G)$ is nontrivial and can be chosen so it corresponds to the basic central extension of $LG$. So $Pic(LG)$ is at least $\mathbb{Z}$ and I think that should be everything.

A divisor for the basic line bundle can be taken to be the complement of $L^-G\cdot L^+G$ where $L^-G = G[z^{-1}]$ so I think the claim would follow if $Pic(L^-G \cdot L^+G) = 0$; but I haven't seen this question addressed in my limited knowledge of the literature.

So is it true that $Pic(LG) = \mathbb{Z}$ and what's a reference that discusses this?


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.