Skip to main content
2 of 4
added 914 characters in body
TCiur
  • 679
  • 3
  • 8

Interpretation of Tate conjecture using motivic homotopy

For a smooth projective variety $X$ over a field $k$ the Tate conjecture says that the cycle class maps $$CH^i(X)\otimes \mathbb{Q}_l \to H^{2i}(X,\mathbb{Q}_l(i))^{G_k}$$

are surjective. To my knowledge both the Chow groups and l-adic cohomology are representable in the motivic category by some ring spectra. Is there a way to interpret the cycle class map, and thus Tate's conjecture, in terms of these ring spectra? I haven't touched algebraic homotopy theory for a long time so this may be a naive question.

I will just add my attempt to understand this so far. In Mixed Weil Cohomologies, Theorem 1 there is a cycle class map from motivic cohomology to any "Mixed Weil Cohomology" $E$ given by $$H^q(X,Q(p)) \to H^q(X,E(p))$$ and from (2.3.24.3) is should arise from a cycle class map on spectra $cl: HQ \to \mathcal{E}$ when $\mathcal{E}$ represents $E$. We know that $H^{2n}(X,Q(n)) = CH^n(X)_Q$. Is $Q$ just an algebra which we can take to be $\mathbb{Q}_l$? In which case we would be done if $l$-adic cohomology was a mixed Weil cohomology.

This is not exactly the case because $l$-adic cohomology is an inverse limit of 'etale cohomology tensored by $\mathbb{Q}_l$. I would be satisfied with using $H^{2i}(X,\mu_l(i))$ instead. Alternatively, we could use Scholze and Bhatt's pro-etale site to define $l$-adic cohomology and then maybe it will be representable?

TCiur
  • 679
  • 3
  • 8