3
$\begingroup$

I knew that two linearly independent and nowhere-vanishing vector fields provide a basis for the tangent space at each point in $\mathbb{R}^{2}$.

Is it necessary that these two vector fields commute? Would you give me an example for these two vector fields?

$\endgroup$
  • 1
    $\begingroup$ What are "these commutators"? No commutators or operators, for that matter, had been mentioned before that sentence. Otherwise. Vector fields $(1,0)$ and $(1,0)$, independent of $x,y$, are surely an example and one can easily write down infinitely many others, pretty much arbitrary fields, right? What is the real question? $\endgroup$ – Luboš Motl Nov 30 '12 at 15:57
  • 3
    $\begingroup$ $\partial/\partial x +f(x,y)\partial/\partial y$ and $\partial/\partial y$ are linearly independent vector fields for any choice of smooth function $f$. We can choose $f$ such that commutators do not vanish. For $f=0$ commutators vanish. $\endgroup$ – user10001 Nov 30 '12 at 16:04
  • $\begingroup$ @Luboš Motl I read somewhere that these fields cannot be a coordinate basis since the commutator does't vanish (like what we have in QM). But I don't know why. $\endgroup$ – Fatima Nov 30 '12 at 16:05
  • $\begingroup$ @Fatima: the commutator of those fields certainly vanishes: $[\partial_{x}, \partial_{y}] = 0$ trivially. $\endgroup$ – Jerry Schirmer Nov 30 '12 at 17:04
  • 1
    $\begingroup$ I think @Fatima might be asking if it is necessary that the Lie derivative $ \mathcal{L}_X(Y) = [X,Y]$ vanishes in order for them to be a coordinate basis? In which case, no, as long as they're non vanishing and independent at each point, you're fine. $\endgroup$ – twistor59 Nov 30 '12 at 18:04
3
$\begingroup$

Consider the vectors $$\partial_{\theta}=-y\partial_{x}+x\partial_{y}$$ and $$\partial_{r}=\frac{x}{\sqrt{x^{2}+y^{2}}}\partial_{x}+\frac{y}{\sqrt{x^{2}+y^{2}}}\partial_{y}$$ Then $$\partial_{\theta}\partial_{r}-\partial_{r}\partial_{\theta}\neq0$$

$\endgroup$
  • 7
    $\begingroup$ Those fields are not nowhere vanishing. $\endgroup$ – Jerry Schirmer Nov 30 '12 at 16:49
  • 1
    $\begingroup$ I think my favourite vector fields would be $\frac{\partial}{\partial x}$ and $\frac{\partial}{\partial y}$ ! $\endgroup$ – twistor59 Nov 30 '12 at 17:59
3
$\begingroup$

Let us take the first vector field to be given by $V_{1}=\partial _{x}$. Any other vector field will be given by $V_{2} = A(x, y)\partial _{x} + B(x, y)\partial _{y}$ . The commutator will be given by

$[V_{1} , V_{2} ] = [\partial _{x} A(x, y)]\partial _{x} + [\partial _{x} B(x, y)]\partial _{y}$

We want this not to vanish. So either one of A, B must depend on x. Set $V_{2} =x \partial _{x} + \partial _{y}$ . $V_{1}$ , $V_{2}$ are nowhere vanishing, and their commutator is nowhere vanishing as well.

$\endgroup$
  • $\begingroup$ Did you mean $V_2=x\partial_x+\partial_y$ ? I mean your $V_1$ and $V_2$ are not independent everywhere (specifically they are not independent at $x=0$). $\endgroup$ – user10001 Nov 30 '12 at 22:04
  • $\begingroup$ @dushya Yes, you are right. $\endgroup$ – Fatima Dec 1 '12 at 20:23
0
$\begingroup$

I think, @Alex answer is not correct. A slight change is required to Alex's answer.

For the vectors, $$\partial_{\theta}=-y\partial_{x}+x\partial_{y}$$ and $$\partial_{r}= x\partial_{x}+y\partial_{y},$$ $$\partial_{\theta}\partial_{r}-\partial_{r}\partial_{\theta}\neq0,$$ whereas for the vectors, $$\partial_{\theta}=-y\partial_{x}+x\partial_{y}$$ and $$\partial_{r}=\frac{x}{\sqrt{x^{2}+y^{2}}}\partial_{x}+\frac{y}{\sqrt{x^{2}+y^{2}}}\partial_{y},$$ $$\partial_{\theta}\partial_{r}-\partial_{r}\partial_{\theta} = 0.$$ Hence, the first set of vectors is the correct answer, not the second.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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