Closed embeddings of manifold/varieties

In Italian, varietà means both manifold and variety. Here is one of the many theorems that exists for both, and which my coauthor and I have used recently in a work in progress.

Theorem We have that f:X → Y  is a closed embedding if and only if it satisfies all the following conditions:
  1. f is injective on points
  2. f is injective on tangent vectors
  3. f is closed (as a map of topological spaces)

Algebraic version (AV): f is a morphism between schemes of finite type over an algebraically closed field k, we only consider k points. 

Differential version (DV): f is a smooth (holomorphic) map between differentiable (complex) manifolds (all manifolds are without boundary). We write k for R or C, respectively.

A closed embedding means an isomorphism to a closed subscheme AV or submanifold DV of Y. 

Note that the conditions indicated are clearly necessary (only if). 

Proof AV Let Z be the schematic closure of the image. Since f is closed, f:X → Z is surjective; by 1. it is bijective, and since it's closed it's a homeomorphism. Thus there is an induced homomorphism of sheaves of algebras \(O_Z\to f_*O_X\) which by definition of schematic closure is injective.
It remains to prove it's surjective, equivalent to prove surjectivity for \(O_Y\to f_*O_X\). This means for all p in X, \(O_{Y,q} \to O_{X,p}\) is surjective where q:=f(p); equivalently, that \(m_q \to m_p\) is surjective (since \(O_{X,p}=k\oplus m_p\)); by Nakayama, that \(m_q/m_q^2 \to m_p/m_p^2\) is surjective, which means by dualizing that \(T_pX \to T_qY\) is injective.

Proof DV Let Z be the image of X; as before Z is closed in Y and \(f:X \to Z\) is a homeomorphism. It is enough to show Z is a manifold of the same dimension of X, since condition 2 then implies that f is locally a diffeomorphism. By assumption, f is locally a closed embedding of manifolds, that is we can cover X by opens U such that there exists a V in Y open with \(f:U \to V\) a closed embedding.
Let X' be the complement of U in X; it is closed, so its image Z' in Y is closed, thus Z' is closed in Z. Let \(W:=V\cap Z\). Since f is bijective, the inverse image of Z' is X', so the inverse image of W is U, and \(f:U \to W\) is a diffeomorphism. By assumption W is a closed submanifold of V. We can cover Y by such opens V, plus the complement of Z; being a closed submanifold is local, this concludes the proof.

Comments

Post a Comment

Popular posts from this blog

Genius doesn’t trample on mastery and leave it in the dust

This was born as a blue sky thread, instigated by someone asserting the opposite of the title. Let me start by introducing context. I am a senior mathematician, which means I’ve spent 40 years living of and in a group which supposedly has geniuses. Indeed some ideas are truly beautiful, and surprising; and some people have more of those, and of wider impact, than the average practitioner.  I  don’t like the word genius, but let’s use it for simplicity to describe such people. They do not stomp on anyone else: in fact, their typical attitude is to found a school, and (with a few asshole exceptions!) seniors recognise their gift, honour it, and support them as far as possible. Such recognition and support are much more likely if the “genius” is a non-disabled, straight-passing white man. Exceptions exist, of course: maths and physics buildings in Cambridge UK are much more accessible than in Oxford UK because Hawking. In general, again, new ideas are celebrated by weaving them i...
Read more

Sheaf of differentials and tangent bunch for a classical scheme

Classical scheme means locally of finite type over a fixed algebraically closed field k. I just made up the name tangent bunch; it should be called tangent abelian cone, but I chose this to highlight the similarity with tangent bundle. A vector bunch E over a scheme X is a scheme E affine over X with a G_m action (for which the map to X is invariant) which is isomorphic to Spec Sym F for some F\in Coh(X). A morphism is a G_m-equivariant morphism of X schemes.  Theorem The functor Spec Sym defines an equivalence of categories Coh(X)^{op} to Vector Bunches on X, which commutes with pullbacks. We know that for a smooth scheme X over k, the individual tangent spaces T_pX neatly organize themselves into a vector bundle TX over X, just as they do for manifolds. But in general, the dimension of T_pX depends on p.  We start with an affine scheme X=Spec R, and write R=k[x,_1,...,x_n]/I, i.e. we choose a closed embedding of X in A^n. We want to define the tangent vector bunch TX to X as...
Read more