3
$\begingroup$

Let $p=2$ or 3, and let $k$ be an algebraically closed field of char. $p.$ Let $E$ be the supersingular elliptic curve over $k$ (with $j=0$). Let $G$ be the automorphism group of $E,$ which has order 12 (resp. 24) when $p=3$ (resp. 2). Then the $\ell$-adic cohomology $H^1(E,\mathbb Q_{\ell})$ is a 2-dimensional representation of $G.$ Is it irreducible?

Since we know the group structure of $G$ (cf. Silverman's AEC, Appendix A, exercise..), does anyone have a reference for its character table?

$\endgroup$
0

2 Answers 2

7
$\begingroup$

If you have a supersingular curve in general, the $\ell$-adic cohomology is a faithful representation of the endomorphism ring, which is a quaternion order that splits over $\mathbb{Q}_\ell$. In the case of characteristic 2 and 3, the automorphism groups are non-abelian (in particular, isomorphic to $SL_2(\mathbb{F}_3)$ and $\mathbb{Z}/3\mathbb{Z} \rtimes \mathbb{Z}/4\mathbb{Z}$, respectively). For non-abelian groups, any faithful two dimensional representation (away from characteristic dividing the order) is automatically irreducible. You can find character tables of the groups in GAP, using commands like:

g := SpecialLinearGroup(2,3);
tbl := CharacterTable(g);
Display(tbl);

In SAGE, you can add "gap" to the beginnning of things:

g = gap.SpecialLinearGroup(2,3)
tbl = gap.CharacterTable(g)
gap.Display(tbl)

There might be a better way to do this.

$\endgroup$
0
7
$\begingroup$

When $p = 2$, the group $G$ is the group of units in the Hurwitz order in the Hamilton quaternions $\mathbb H$ over $\mathbb Q$, i.e. the group $\{\pm 1,\pm i,\pm j,\pm k,(\pm 1 \pm i \pm j\pm k)/2\}$. It is a central extension of $A_4$ by a group of order $2$.

The $\ell$-adic cohomology is given by the standard two-dimensional representation of $M_2(\mathbb Q_{\ell}) := \mathbb H \otimes_{\mathbb Q} \mathbb Q_{\ell}$. This is an irrep. of $G$.

I can't write down the $p = 3$ case off the top of my head like this, but it will be similar, and the representation on $H^1(\mathbb Q_{\ell})$ should again be irreducible.

$\endgroup$

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.