Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

Let $S^2$ be the 2-sphere in $\mathbb{R}^3$, given with vector field $X=x \partial_x+ y \partial_y$ on the stereographic projection from north pole chart. What is the global extension of $X$ to $S^2$ as a vector field on $\mathbb{R}^3$?

Edit: The same vector field $X=x \partial_x+ y \partial_y$ on south pole chart(I checked that on overlap these define same vector fields) gives a globally defined vector field on sphere. I think this problem wants to ask how can $X$ be expressed in terms of $x,y,z$ in $\mathbb{R}^3$. (Since $x,y$ in $X$ is not the same $x,y$ in $\mathbb{R}^3$.)

share|cite|improve this question
You've defined $X$ at every point of $\mathbb{S}^2$ except the north pole. What value does $X$ need to take in $T_{\mbox{north pole}}\mathbb{S}^2$ so that it is smooth there? – Neal Dec 12 '12 at 15:37
$\mathbb{R}^3$ is completely redundant here. – Rhys Dec 12 '12 at 16:05
I made some edits to clarify things. – Gobi Dec 12 '12 at 16:13
Dear Gobi,if I correctly interpret what you say, there is a change of sign in the expression of the field in the south pole chart: see my answer. – Georges Elencwajg Dec 23 '12 at 11:49

As far as I know, there is no canonical way to extend vector fields from a submanifold. The field can be extended in many ways, for example by using a smooth radial cut-off function $\eta$ that is equal to $1$ in a neighborhood of $S^2$ and to $0$ at some distance from $S^2$. For example, if the original field was $\sin \phi \frac{d}{d\phi}$ in spherical coordinates $(\theta,\phi)$ on $S^2$, then $\eta(\rho)\sin \phi \frac{d}{d\phi}$ would be its extension to $\mathbb R^3$ in spherical coordinates $(\rho, \theta,\phi)$.

share|cite|improve this answer

The change of coordinates from north pole coordinates to south pole oordinates is $$\phi:\mathbb R^2\setminus \lbrace (0,0)\rbrace \to \mathbb R^2\setminus \lbrace (0,0)\rbrace:\begin{pmatrix} x \\\\ y\end{pmatrix}\mapsto \begin{pmatrix} u \\\\ v\end{pmatrix}=\frac {1}{(x^2+y^2)}\begin{pmatrix} x \\\\ y\end{pmatrix}$$ and the jacobian of that map is given by the matrix

$$Jac(\phi)(x,y)=\frac {1}{(x^2+y^2)^2}\begin{pmatrix} -x^2+y^2&-2xy \\\\ -2xy&x^2-y^2 \end{pmatrix}$$

This implies that your vector field has as expression in the south pole chart(beware the minus sign!) $$Y(u,v)=-(u\partial_u+v\partial_v)\quad \text{for} \begin{pmatrix} u \\\\ v\end{pmatrix}\in \mathbb R^2\setminus \lbrace (0,0)\rbrace $$ which can be smoothly extended to the origin by $Y(0,0)=0\partial_u+0\partial_v$.

The initial vector field $X$ on the sphere minus the north pole smoothly extends to a global vector field on the whole sphere by defining it as zero at the north pole.

Slightly more generally we can define a global vector field (vanishing at the two poles and only there if the real numbers $a,b$ are not both zeroon the sphere by the following expressions in the two charts given by the two stereographic projections: $$(ax+by)\partial_x+(-bx+ay) \partial_y=(-au+bv)\partial_u+(-bu-av)\partial_v$$ If the real numbers $a,b $ are not both zero, that vector field vanishes at the two poles and only there.

share|cite|improve this answer

Your Answer


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

Not the answer you're looking for? Browse other questions tagged or ask your own question.