$\DeclareMathOperator{\h}{H}$Apologies in advance if this is overly stupid. Let $k$ be a field and $X$ a variety over $k$. Let $n$ be an integer which is invertible in $k$. One often looks at the groups $\h^i(X_{\bar k},\mathbb Z/n)$, i.e. the cohomology of the constant sheaf $\mathbb Z/n$ on the etale site of $X_{\bar k} = X\times_{\operatorname{Spec} k}\operatorname{Spec}{\bar k}$. We give this a $G_k=\operatorname{Gal}(\bar k/k)$-action via the action of $G_k$ on $\bar k$. I'm wondering if there is a simpler way to do this. The category of continuous $G_k$-modules is equivalent to the category of etale sheaves on $\operatorname{Spec} k$. If we let $f:X\to \operatorname{Spec} k$ be the structure morphism, then for any sheaf $\mathscr F$ on $X$, we have $G_k$-modules $\mathsf R^i f_\ast \mathscr F$. My question is:

Is $\mathsf R^i f_\ast \mathscr F$ naturally isomorphic (as a $G_k$-module) to $\h^i(X_{\bar k}, \mathscr F)$?


This is true if $X/k$ is proper (SGA 4.5, Les points de départ, IV. I, Théorème 1.1).

  • $\begingroup$ This is exactly what I was looking for. Thanks! $\endgroup$ – Daniel Miller Dec 3 '13 at 2:16
  • 1
    $\begingroup$ @DanielMiller My pleasure! It is not a trivial theorem. :) $\endgroup$ – Bruno Joyal Dec 3 '13 at 2:17

Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.