3
$\begingroup$

Let be $\phi: F \rightarrow G$ a morphism of sheaves and $\phi_P:F_P \rightarrow G_P$ the induced map of germs. I proved that for all $P$, $(ker \phi)_P=ker(\phi_P)$ and $(im \phi)_P= im(\phi_P)$. Now I have to show that $\phi$ is injective (resp. surjective) iff the induced map $\phi_P$ is injective (resp. surjective). How can I show that if $\phi_P$ is surjective (injective) than $\phi$ is surjective (injective)?

$\endgroup$
2
$\begingroup$

Assuming what you already have, this seems easy unless I am missing something: If $\phi_P$ is surjective everywhere, let $\phi:F\to G$ factor as $\newcommand{\im}{\mathrm{im}}\chi:F\to\im(\phi)$ and $\psi:\im(\phi)\to G$. The latter map is an isomorphism on stalks, because $\im(\phi)_P=\im(\phi_P)\cong G_P$. Hence, $\im(\phi)\cong G$. On the other hand if $\phi_P$ is injective, then $\ker(\phi)_P=\ker(\phi_P)=0$ so $\ker(\phi)$ is the zero sheaf.

$\endgroup$

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.