Chromatic aberrations of geometric Satake over the regular locus

Abstract

Let G be a connected, simply-laced, almost simple algebraic group over C, let Gc be a maximal compact subgroup of G(C), and let Tc be a maximal torus therein. Let GrG denote the affine Grassmannian of G, and let G denote the Langlands dual group to G with Lie algebra g. The derived geometric Satake equivalence of Bezrukavnikov-Finkelberg gives an equivalence between the ∞-category LocGc(GrG; C) of Gc-equivariant local systems of C-vector spaces on GrG and the ∞-category of quasicoherent sheaves on a large open substack of g[2]/G. In this article, we study the analogous story when LocGc(GrG; C) is replaced by the ∞-category of Tc-equivariant local systems of k-modules over GrG(C), where k is (2-periodic) rational cohomology, (complex) K-theory, or elliptic cohomology. Crucial to our work is the genuine equivariant refinement of these cohomology theories. We show that, although there may not be an equivalence as in derived geometric Satake, the ∞-category LocTc(GrG; k) admits a 1-parameter degeneration to an ∞-category of quasicoherent sheaves built out of the geometry of various Langlands-dual stacks associated to k and the 1-dimensional group scheme computing S1-equivariant k-cohomology. For example, when k is an elliptic cohomology theory with elliptic curve E, the ∞-category LocTc(GrG; k) degenerates to the ∞-category of quasicoherent sheaves on a large open locus in the moduli stack of B-bundles of degree 0 on E. We also study several applications of these equivalences.

0

Turn this paper into a lesson

ArcXiv compiles a structured reading guide from this paper's metadata: plain-English importance, contributions, prerequisite concepts, which sections to read first, flashcards, and a quiz. Grounded in the abstract, never invented.

Discussion (0)

Sign in to join the discussion.

Loading comments…