I have an elliptic curve $E$ defined over a complete discrete-valued field $K$ of characteristc $0$. the residue field $k$ is of positive characteristic $p$.

Then $E[p]=\mathbb{Z}/p\mathbb{Z} \times \mathbb{Z}/p\mathbb{Z}$.

What can I say about the height of the formal group associated to $E$?

Seems that there isn't a definition of height of formal group in characteristic $0$.

However I have another question, the formal group law $F$ for $E$ has the coefficients in $\mathcal{O}_K$ so we can consider their reduction modulo the maximal ideal of $\mathcal{O}_K$. Now assume that $E$ has a good reduction $\tilde E$. Is the formal series obtained by reduction the coefficients of $F$ a formal group law on $\tilde E$ ?

  • $\begingroup$ This isn't possible. The group scheme $E[p]$ is never étale, and so cannot be constant. In more simple terms, if $E[p]$ is as stated, then $[p]:E_k\to E_k$ would be separable, but it's not. $\endgroup$ Aug 11, 2015 at 10:34
  • $\begingroup$ is the height of a formal group defined only in positive characteristic? $\endgroup$
    – user261123
    Aug 11, 2015 at 11:10
  • $\begingroup$ As far as I know, yes. Usually one defines the height of a formal group $\mathbb{G}$ to be the $h$ such that $[p]$ is a powe series in $T^{p^h}$. The proof that such a thing exists (besides for the additive group) uses the Frobenius. What even does the height mean if you're not in positive characteristic? Regardless, you can't have an elliptic scheme over $\mathbb{Z}_p$ with $E[p]$ isomorphic to $\underline{\mathbb{Z}/p\mathbb{Z}}\times\underline{\mathbb{Z}/p\mathbb{Z}}$. $\endgroup$ Aug 11, 2015 at 11:21
  • $\begingroup$ I know that in characteristic $p$ the $p$-torsion points group is not of this form but $K$ is a field of characteristic $0$. I got your reasoning about the height thanks! $\endgroup$
    – user261123
    Aug 11, 2015 at 11:25
  • $\begingroup$ Oh, I apologize. I thought that you were assuming that you had $E/\mathcal{O}_K$—because I thought you were reducing $E$ over $k$. I see now that you were asking about some notion of 'height' over $K$, but yeah, I don't think that exists. $\endgroup$ Aug 11, 2015 at 11:32

1 Answer 1


If a formal group is defined over a $p$-adic ring $\mathscr O$, with residue field $k$ of characteristic $p$, then there certainly is a pretty standard definition of the height of the formal group. It’s just the height of the $k$-series. All of this is in the very old text of Fröhlich, I think. As you expect, the height of the formal group of an elliptic curve is $1$, $2$, or $\infty$, but always finite if the reduction is good.

I’m not sure about your second question. What is “a formal group on $\tilde E\,$” ? It’s certainly true, though, that the processes (1) of taking reduction modulo the maximal ideal and (2) localization to get the formal group commute with each other.


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .