8
$\begingroup$

Let $X$ be an algebraic stack of finite type over a field.

Is there an intrinsic way to calculate the minimum of the dimensions of all atlases of $X$?

By intrinsic here I mean using constructions such as the inertia stack, the stabilizer group construction, etc.

A natural conjecture is that this minimum should be something like the dimension of $X$ plus the dimension of the largest stabilizer of any point of $X$, as one can see using classifying stacks, for example.

$\endgroup$

2 Answers 2

3
$\begingroup$

This is false: $B\mu_p$ in characteristic $p$ is an algebraic stack but not a DM stack (and in particular, does not have a 0-dimensional smooth cover, because such a cover would be etale).

(Note that while $B\mu_p$ is a point mod $\mu_p$, the map from the point to $B\mu_p$ is a $\mu_p$ torsor, and thus not etale since $\mu_p$ is not etale. It does have a cover by $\mathbb{G}_m$.)

$\endgroup$
1
  • 1
    $\begingroup$ Maybe one could define a smooth envelope for a group scheme as a smooth group scheme into which it embeds, so that the smooth envelope of $\mu_p$ is $\mathbb{G}_m$? Then it might be the dimension of $X$ plus the largest dimension of a smooth envelope of a stabilizer. $\endgroup$ Jan 10, 2023 at 10:46
1
$\begingroup$

Professor Jarod Alper provided me an answer to this question in separate correspondence. When the stabilizer at a point of the stack is smooth, my conjecture above is true by the theory of miniversal deformations. This is Theorem 3.6.1 in Professor Alper's notes here: https://sites.math.washington.edu/~jarod/moduli.pdf.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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