Proving the Differential Map (Pushforward) is Well-Defined

In summary, the conversation discusses proving the well-definition of the differential map (pushforward) in a graduate math course. The map is defined as a smooth map between two smooth manifolds, and it is necessary to show that it is surjective and does not depend on a certain choice of chart or other parameters. The group also discusses choosing a definition for the differential map and ensuring that it is well-defined.
  • #1
Fgard
15
1
I am taking my first graduate math course and I am not really familiar with the thought process. My professor told us to think about how to prove that the differential map (pushforward) is well-defined.

The map
$$f:M\rightarrow N$$ is a smooth map, where ##M, N## are two smooth manifolds. If ##(U,x)\in A_M## and ##(V,y)\in A_N## are two charts in their respective atlas then the map ##x\circ f\circ y^{-1} ## is also smooth. Then what remains to prove is that ## f## is surjective. Is this correct? Or do I need to prove something else as well?
 
Physics news on Phys.org
  • #2
No, what you have is the map, the differential of this map is another map, that takes tangent vectors on ##M## to tangent vectors on ##N##. You have to prove that whatever definition was given in the lectures (something we can only guess) is well defined.
 
  • Like
Likes lavinia and Orodruin
  • #3
Fgard said:
I am taking my first graduate math course and I am not really familiar with the thought process. My professor told us to think about how to prove that the differential map (pushforward) is well-defined.

The map
$$f:M\rightarrow N$$ is a smooth map, where ##M, N## are two smooth manifolds. If ##(U,x)\in A_M## and ##(V,y)\in A_N## are two charts in their respective atlas then the map ##x\circ f\circ y^{-1} ## is also smooth. Then what remains to prove is that ## f## is surjective. Is this correct? Or do I need to prove something else as well?
Shouldn't it be ##y\circ f\circ x^{-1} ## which is smooth? But besides this, the differentiability is given by definition. If the chart mappings weren't smooth, we wouldn't call the manifolds smooth. Also surjectivity isn't needed except you only want to consider ##N=f(M)##. Well-definition means, that no two elements of ##N## can be the image of one point in ##M##. What does this mean in the given situation? We don't have an element in ##M## that could end up in two different ways under ##f##. But we do have eventually two different charts for this element ##x \in M##. Could it end up in two different charts of ##N## where they represent different points?
 
  • #4
Okej, so I have to choose a definition for the differential map and show that map dose not depend on a certain choice of chart. Thanks.
 
  • #5
Fgard said:
Okej, so I have to choose a definition for the differential map and show that map dose not depend on a certain choice of chart. Thanks.
Well, you should choose the one your professor told you to think about. And you need to show it doesn't depend on whatever choice was made, it could be a chart, but it could be something else. For example given a vector on M, choose a curve, whose tangent vector is the given one, then map the curve to N, take the tangent vector. If your definition is something along these lines, then you need to show that it doesn't depend on the choice of the curve.
 

Similar threads

Replies
10
Views
2K
Replies
20
Views
3K
Replies
36
Views
3K
Replies
3
Views
1K
Replies
12
Views
4K
Replies
10
Views
3K
Replies
3
Views
2K
Replies
5
Views
3K
Back
Top