Hello,
Is there a proof that the push forward by a proper morphism of Noetherian schemes sends coherent sheaves to coherent ones, without passing in the argument through projective morphisms?
Thank you, Sasha
$\begingroup$
$\endgroup$
4
-
$\begingroup$ Well, you can use Chow's lemma, the result for projective morphisms, and the Leray spectral sequence to get it. Or you can find a different proof here EGA III 3.2.1 here: numdam.org/numdam-bin/feuilleter?id=PMIHES_1961__11_ $\endgroup$– ParsaCommented Dec 6, 2011 at 15:38
-
1$\begingroup$ @Parsa: If you use Chow's lemma, you are using the result for projective morphisms, which is what Sasha wants to avoid? $\endgroup$– user19475Commented Dec 6, 2011 at 17:22
-
$\begingroup$ If your schemes are locally of finite type over a field k, Antoine Ducros gave a proof that meets your requirements (see arXiv:math/0612521). Basically, the result is known for proper rigid analytic spaces (it is a theorem of Kiehl), so one also gets it for analytifications of proper morphisms (analytifications are to be taken in Berkovich sense, over k endowed with the trivial absolute value) and for proper morphisms themselves. $\endgroup$– Jérôme PoineauCommented Dec 6, 2011 at 22:35
-
$\begingroup$ @Timo: I wasn't sure what he meant by "without passing in the argument through projective morphisms" so I gave an approach assuming proper but not projective (allowing the result for projective) and referenced another proof using only proper. $\endgroup$– ParsaCommented Dec 7, 2011 at 0:21
Add a comment
|
1 Answer
$\begingroup$
$\endgroup$
0
Gerd Faltings, Finiteness of coherent cohomology for proper fppf stacks, J. Algebraic Geometry 12 (2003) 357–366
