Homepage › Solution manuals › John Lee › Introduction to Smooth Manifolds › Problem 8-13 (A smooth vector field on the sphere vanishing at a single point)
Problem 8-13 (A smooth vector field on the sphere vanishing at a single point)
Show that there is a smooth vector field on that vanishes at exactly one point. [Hint: try using stereographic projection; see Problem 1-7.]
Answers
First, we introduce our intuition behind the solution:
-
A smooth chart, being a diffeomorphism, determines a tangent vector to each point of the given manifold through the differential of its inverse. More precisely, if is any smooth chart on , the assignment
determines a vector field on . It is smooth because its component functions are constant.
-
Stereographic projection given by
is an example of such smooth chart.
If we visualise both and its projection on on the same graph, we see that the mapping chart gets flatter and flatter towards the north pole - but not completely flat.
- Solution: Define the desired vector space by associating it with the differential of (in the standard coordinates); set the tangent at the north pole to be zero.
We demonstrate that this strategy gives the desired result.
Proof. Define our vector field by
- 1.
-
does not vanish on
Since is a vector space homomorphism for all , it will not map the standard basis derivations on to zero derivations on . In other words, cannot vanish anywhere on . - 2.
-
is smooth
Recall Proposition 8.1 (Smoothness Criterion for Vector Fields), which states that given a smooth chart , the restriction of to is smooth if and only if its component functions with respect to this chart are smooth1 . This, by construction, makes automatically smooth on . Thus, it is only left to demonstrate that is smooth at . - 3.
-
is smooth at the north pole
Consider the second chart in our atlas which covers the north pole, i.e., the south pole projection , given byThe change of coordinates is given by
We compute the coordinate representation of in terms of the smooth chart :
where denotes the coordinates with respect to the south pole projections.
Since , we have
We see that the component functions are smooth.
In other words, the representation agrees on and is smooth on since the component functions are zero at . Thus, is smooth at .
Comments
-
Should be an error in the equation after "The change of coordinates..." in part 3. The coeffecient in the second last line reads x^2 -1 / x^2 - 1 when it should be x^2 - 1 / x^2 + 1, unless I'm mistaken.macquarrielucas • 2024-01-10