3
$\begingroup$

Im facing a problem in Silvermans Book "Arithmetic of elliptic Curves" at the beginning of chapter X.4 concerning the exact sequences.

Let $K$ be a number field with a valutaion $v$. I'm distinguishing between the algebraic closure of $K_v$, which i denote as $\overline{K_v}$ and the completion of the algebraic closure of $K$, which is defined as $L = \bigcup L_{iv}$ with the $L_i$ ranging over all finite extensions of $K$ (as done in Neukirch). Now to obtain the short exact sequence $(\ast)_v$ used to define the Selmer-Group in Silverman's "Arithmetic of Elliptic Curves", page 331:

$ 0 \longrightarrow E'(K_v)/\phi E(K_v) \longrightarrow H^1 (G_v, E[\phi]) \longrightarrow H^1 (G_v, E)[\phi] \longrightarrow 0 $

I have to work with $\overline{K_v}$, as I need an algebraically closed field. However, how can one let $G_v\subset Gal( \overline{K} | K)$ act on $\overline{K_v}$? I know $G_v \simeq Gal(L | K_v)$, but do not see, how to use that isomorphism here. On the other side, if I work with $L$, I don't know how to identify $H^1 (G_v, E)$ with the Weil-Chatelet-Group $WC(E/K_v)$, which is in bijection to $H^1 (Gal(\overline{K_v}|K_v), E)$.

Im very thankful for every advice given!


For those of you not having Silverman's book at your hands, it can be found here http://www.mathe2.uni-bayreuth.de/stoll/talks/short-course-descent.pdf. The section described above begins on page 331.

$\endgroup$
1
$\begingroup$

The two fields you are considering are the same, because of Krasner's lemma, which has as a corollary that separable closure and $p$-adic completion commute for global fields. See for example prop. 8.1.5 of Neukirch's "Cohomology of number fields".

$\endgroup$
  • $\begingroup$ I've read, the algebraic closure of $\mathbb{Q}_p$ is not complete. How is this consistent with that proposition? $\endgroup$ – NeukirchLover Aug 5 '15 at 15:21
  • $\begingroup$ I think the field L im considering above is not complete, so this is consistent with the non-completeness of the algebraic closure of the p-adic numbers. $\endgroup$ – NeukirchLover Aug 6 '15 at 8:17

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.