Local Tate duality for syntomic cohomology in positive characteristic

summary

The gist

Let k be a local field of characteristic p > 0.

In short

The paper establishes local Tate duality for syntomic cohomology over a local field k of characteristic p > 0 using Dodd’s theory of F-gauges. It shows this cohomology satisfies duality when defined via F-gauges and is Lagrangian over the ring of integers. For p > 2, these results recover Shatz’s and Milne’s duality theorems for flat group schemes, connecting syntomic cohomology to geometric objects.

Key concepts

F-gauges
These are derived p-complete complexes of graded left Db(m−1,m),⋆C-modules. They are used as the correct setting for studying duality in p-adic cohomology. They can be refined by conditions like being lisse or quasi-lisse, and their category is symmetric monoidal.
Syntomic Cohomology
This is a cohomology group associated with an F-gauge M. For a smooth variety, it computes coherent cohomology on k Syn. The paper shows it relates to the cohomology of virtual F-crystals and finite flat group schemes of p-power order over k.
Local Tate Duality
This is a fundamental duality result showing an equivalence between a complex and its dual in the derived category D(Zp). The paper proves this holds for syntomic cohomology when defined using F-gauges, linking it to Pontryagin duality for vector spaces.
Lagrangian Refinement
This refinement shows that the duality pairing over the ring of integers is not just an equivalence but involves a fibre sequence. This is achieved by showing that certain induced pairings are zero for lisse F-gauges, leading to a structural result involving Nygaard and de Rham cohomology.

Terminology used across episodes

This episode discusses

The paper

Local Tate duality for syntomic cohomology in positive characteristic · Read on arXiv

Let k be a local field of characteristic p>0. We show that syntomic cohomology over k, with coefficients, and defined using Dodd's theory of F-gauges, satisfies local Tate duality. We also prove that syntomic cohomology over the ring of integers of k, again with coefficients, is Lagrangian with respect to the duality pairing. Finally, we show that, for p>2, these results recover Shatz's duality and Milne's orthogonality results for the cohomology of finite flat group schemes of p-power order over k.

Transcript

Introduction to the show: ident: Mathematics. The day in mathematics: new theorems, new proofs, and what they open.

Ada: I'm Ada.

Emmy: And I'm Emmy, here with Kolya.

Ada: We're looking at the paper: "Local Tate duality for syntomic cohomology in positive characteristic".

Emmy: Here's the gist of the paper. Let k be a local field of characteristic p > 0.

Ada: We show that syntomic cohomology over k, with coefficients, and defined using Dodd’s theory of F-gauges, satisfies local Tate duality for all i.

Emmy: We also prove that syntomic cohomology over the ring of integers of k, again with coefficients, is Lagrangian with respect to the duality pairing.

Ada: Finally, we show that, for p > 2, these results recover Shatz’s duality and Milne’s orthogonality results for the cohomology of finite flat group schemes of p-power order over k.

Paper summary: Ada: So this paper, "Local Tate duality for syntomic cohomology in positive characteristic," it's tackling the idea of local Tate duality over a local field k with characteristic p greater than zero. The main thesis is that syntomic cohomology, when you define it using Dodd’s theory of F-gauges, actually satisfies this kind of duality.

Emmy: And they go further by proving that when you look at syntomic cohomology over the ring of integers of k, with coefficients, it's Lagrangian with respect to the duality pairing. It also shows that for p greater than two, these results connect to Shatz’s duality and Milne’s orthogonality results for the cohomology of finite flat group schemes.

Kolya: So what's the big picture here? They start by recalling local Tate duality for discrete Gk modules, but they run into trouble in positive characteristic because there isn't p-torsion in that module, which messes up the naive setup. The paper suggests using a different setting—the category of perfect complexes on the syntomification k Syn—as the right place to get these duality results in p-adic cohomology.

Ada: Exactly. They introduce F-gauges as their main tool, defined as derived p-complete complexes of graded left Db(m minus one, m) star C modules. These gauges are then refined by being lisse or quasi-lisse, and the category of F-gauges has a nice symmetric monoidal structure using the derived tensor product.

Emmy: And they show that this theory extends to gauges of any level m, and that the independence of these categories from choosing a Cohen ring C and a Frobenius lift sigma is shown under certain conditions on m. That’s important for making the construction concrete.

Kolya: So when they get to computing cohomology groups, they define Nygaard cohomology, de Rham cohomology with level zero, and syntomic cohomology itself. They give explicit formulas for these groups using double complexes involving the module of differentials and the gauge components.

Ada: And then they draw a comparison between F-gauges and other things like virtual F-crystals or p-divisible groups. They show that syntomic cohomology computes what you'd expect for those objects, like RGammaSyn(k, g(zero)M) being represented by the totalisation of a double complex involving divided Frobenius maps <ref:2610.21622#pg1>.

Emmy: And they connect finite flat group schemes of p-power order to F-gauges. They establish a natural equivalence between the cohomology of these group schemes and that of an F-gauge. This really lets them translate geometric objects into this more computable framework.

Conclusion: Ada: So, to wrap up, the paper "Local Tate duality for syntomic cohomology in positive characteristic" shows that we can establish local Tate duality for syntomic cohomology over a local field k of characteristic p greater than zero when defined via Dodd’s theory of F-gauges.

Emmy: And they also prove the Lagrangian refinement of this duality over the ring of integers of k with respect to the duality pairing. This means for a lisse F-gauge M over k with coefficients, the induced pairing is zero, leading to a fibre square that turns out to be a fibre sequence.

Kolya: So what does this mean practically? It connects the abstract theory back to things we know from other areas like Shatz’s duality and Milne’s orthogonality results for finite flat group schemes of p-power order.

Ada: It suggests that syntomic cohomology is a really good, robust tool for studying these objects in positive characteristic, especially when you use this F-gauge machinery. It gives us a concrete way to compute things that were previously harder to handle due to the lack of p-torsion issues.

Emmy: The whole point is establishing this duality and its refinement through the Lagrangian structure, which ultimately boils down to Pontryagin duality for finitely generated k vector spaces, as they show.

Kolya: So if you're looking at p-adic cohomology in positive characteristic, this paper gives you a framework where you can actually get those dualities working by relating it to these F-gauges. That’s the main utility here.

More episodes

← Home