Only the prime two can ramify. More precisely, every field is unramified at every odd prime; actual ramification at two depends on and the chosen division point. Use , as printed in the original PDF.
Fix an odd prime and put . The elliptic-curve discriminant gives good reduction at , and is a unit in . The unramified division torsors at good primes imply that all division points of an elliptic curve above are defined over finite unramified extensions of . Here is a direct lifting proof, which does not require to have integral affine coordinates.
Properness gives a reduction . Choose a point with over the algebraic closure of the residue field, and a finite residue extension containing it. Pass to the corresponding finite unramified extension . The surjectivity of good reduction over a local field lifts to . Its error lies in the kernel of reduction of an elliptic curve. The prime-to-residue-characteristic multiplication on a formal group makes bijective on that kernel. Choose the unique kernel point with ; then satisfies and has reduction .
Every algebraic division point has some such reduction and is the unique lift in its division fibre: the difference of two lifts lies in the reduction kernel, which has no -torsion. Thus every chosen lies in an unramified local extension. For every prime of its possibly non-Galois coordinate field above , the ramification index is one. Globally,One cannot universally assert that two actually ramifies: and give the field . It can ramify. For example, a half of isThe elliptic curve group law verifies this by doubling. Its coordinate field contains , which has ramification in the quadratic extension generated by the square root of two, so the only possible bad prime does occur for this choice.
Articles by others on the same topic
There are currently no matching articles.