MathOverflow is a question and answer site for professional mathematicians. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

This might be a naive question. But since I haven't seen this in any reference, I'll try to ask it here. Let $T$ be a smooth scheme over the algebraically closed field $k$ of characteristic $p>0$ (we can assume that $T$ is affine). Let $\mathbb{T}$ be a smooth lift of $T$ over the ring $W_{2}(k)$, the ring of witt vectors of length 2 over k. Now, if $H_{\mathbb{T}}$ is an F-crystal on $\mathbb{T}$ and $F_{1}$ and $F_{2}$ are two lifts of Frobenius $F_{T}$ on T to $\mathbb{T}$, I know that there must exist an isomorphism $X(F_{1}, F_{2})$ between $F_{1}^*H_{\mathbb{T}}$ and $F_{2}^*H_{\mathbb{T}}$ coming from the nconnection on $H_{\mathbb{T}}$. What is the explicit formula of this isomorphism?

share|cite|improve this question
up vote 4 down vote accepted

This is parallel transport. Note that you only get an explicit formula for the isomorphism between $F_1^*H(\mathbb{T})$ and $F_2^*H(\mathbb{T})$. To get explicit formulas over other thickenings of $\mathbb{T}$, you would need lifts of both $\mathbb{T}$ and also of the Frobenius lifts. For convenience, set $H_1=H(\mathbb{T})$. We will assume that $\mathbb{T}$ has \'etale co-ordinates $x_1,\ldots,x_n$.

Since $H_{\mathbb{T}}$ is a crystal over $\mathbb{T}$, there is a flat connection $\nabla:H_1\to H_1\otimes\Omega^1_{\mathbb{T}/W_2(k)}$. Note that, for any $a\in\mathcal{O}_{\mathbb{T}}$, $\delta(a)=F_1(a)-F_2(a)$ is in $p\mathcal{O}_{\mathbb{T}}$; in particular, $\delta(a)^2=0$. Now, for any $h\in H_1$, the isomorphism sends $F_1^*h$ to $F_2^*h+\sum_{i=1}^n\nabla(\partial_i)(m)\delta(x_i)$. This is just a Taylor expansion up to the first degree.

share|cite|improve this answer
Dear Keerthi. Thank you for your answer. what do you mean by $\nabla(\partial_i)(m)$ ? I think you really mean $\nabla(h)(\partial_i)$. Am I right? – Jack Sep 29 '11 at 13:30
Sorry, $m$ should be $h$; and I suppose $h$ and $\partial_i$ commute in this notation! – Keerthi Madapusi Pera Sep 29 '11 at 13:46

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

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