Take the 2-minute tour ×
MathOverflow is a question and answer site for professional mathematicians. It's 100% free, no registration required.

Let $M$ be a smooth compact, oriented manifold. Let $X$ be a submanifold which is of the following type $$X := \{ p \in M: \psi(p) =0, ~~\varphi(p) \neq 0 \} $$ where $$ \psi: M \rightarrow V, \qquad \varphi: M \rightarrow W $$ are smooth sections of some oriented vector bundle $V$ and $W$. Moreover, when $\varphi(p) \neq 0$, the section $\psi$ is transverse to the zero set. In particular $X$ is a smooth oriented manifold of the right dimension. My question is the following: Given a smooth oriented vector bundle $E \rightarrow M$ such that rank of $E$ is same as dimension of $X$, is it always possible to find a smooth section $s:M \rightarrow E$, transverse to the zero set, such that $s^{-1}(0)$ intersects $X$ transversally and it does not intersect $\overline{X}-X$ anywhere, where $\overline{X}$ is the closure of $X$ inside $M$?

Secondly, is it necessarily true that the object $\overline{X}$ defines a homology class in $H_{*}(M,\mathbb{Z})$? In particular, I want to say that $$ < e(E), [\overline{X}] > = +-| X \cap s^{-1}(0) | $$

where $e(E)$ is the Euler class of $E$ and $+-|A|$ indicates the signed cardinality of the set $A$.

Note that, if $X$ was simply some smooth submanifold of $M$ then this may not have been true, because the closure could be a very large set. The point here is that $X$ is given as the vanishing and nonvanishing of some sections, which are defined on the whole of $M$. If instead $\psi$ was defined only partially on $M$ (eg a section with poles), then the statement might not have been true.

share|improve this question
add comment

Your Answer

 
discard

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

Browse other questions tagged or ask your own question.