6
$\begingroup$

I'm trying to compute $\int_M \omega$ with \begin{align*} \omega &= x^4 dy \wedge dz + y^4 dz \wedge dx + z^4 dx \wedge dy, \\ M &: x^2 + y^2 + z^2 = R^2. \end{align*} I have done this in two ways: Stokes' theorem and direct computation of the wedge product via a spherical coordinate transformation. Using $\phi$ as azimuthal and $\theta$ as polar angle and applying Stokes I obtain \begin{align*} 4 \int_0^{2 \pi}\int_0^{\pi} (x^3 + y^3 + z^3) R^2 \sin \theta d\theta d\phi = 0, \end{align*} where I used the standard \begin{align*} x &= R \sin \theta \cos \phi \\ y &= R \sin \theta \sin \phi \\ z &= R \cos \theta. \end{align*} By directly computing the wedge products and plugging them in I obtain \begin{align*} dy \wedge dz &= R^2 \sin^2 \theta \cos \phi d\theta \wedge d\phi \\ dz \wedge dx &= R^2 \sin^2 \theta \sin \phi d\theta \wedge d\phi \\ dx \wedge dy &= R^2 \cos \theta \sin \theta d\theta \wedge d\phi \\ \int_M x^4 dy &\wedge dz + y^4 dz \wedge dx + z^4 dx \wedge dy = 0. \end{align*} The solution given by my professor was $\frac{12}{5} \pi^2 R^5$, which neither method agrees with. Then I repeat this procedure but instead of taking a sphere, I take a hemisphere. Changing the limits in the $\theta$ integrals to $[0, \frac{\pi}{2}]$, the Stokes' method yields $2 \pi R^5$ and the direct wedge computation yields $\frac{\pi R^6}{3}$. The solution given then is just $\frac{6}{5} \pi^2 R^5$, which also does not agree with my solutions. Any help with my mistakes would be greatly appreciated.

$\endgroup$
4
  • 1
    $\begingroup$ Wolfram mathematica says that the integral, using Stokes, is zero. Maybe there is a typo in the exercise or the order in the differential forms $\endgroup$
    – Masacroso
    Apr 21 at 18:30
  • $\begingroup$ @MasacrosoYeah that is exactly the problem I had, but since the hemisphere's don't give the same solution for both methods I was unsure whether the approach was wrong $\endgroup$
    – Chris
    Apr 21 at 18:37
  • $\begingroup$ note that for a hemisphere we need to add a circular surface to complete the boundary, did you take that into account when using Stoke's theorem? $\endgroup$
    – Masacroso
    Apr 21 at 19:07
  • $\begingroup$ $\frac{\pi R^6}{3}$ is the correct answer and both methods yield the same result. You may have made a mistake to get $2\pi R^5$. Please see my working. $\endgroup$
    – Math Lover
    Apr 21 at 19:34
3
$\begingroup$

In case of sphere, the surface integral is indeed zero.

Please note in below working of mine, I use $\theta$ as azimuthal angle and $\phi$ as polar. I think you have the other way round. So want to call out to avoid confusion.

In case of hemisphere above $z = 0$,

Parametrization of sphere is $r(\phi, \theta) = (R \cos \theta \sin \phi, R \sin\theta \sin\phi, R \cos\phi)$

For outward normal vector, $\displaystyle r'_{\phi} \times r'_\theta = R^2 \sin\phi(\cos\theta \sin\phi, \sin\theta \sin\phi, \cos\phi)$

The integral becomes,

$\displaystyle \small \int_0^{2\pi}\int_0^{\pi/2} R^2 \sin\phi \ (R^4 \cos^4\theta\sin^4\phi, R^4 \sin^4\theta\sin^4\phi, R^4 \cos^4\phi) \cdot (\cos\theta \sin\phi, \sin\theta \sin\phi, \cos\phi) \ d\phi \ d\theta$

$= \displaystyle \frac{\pi R^6}{3} \ $, which is same as your working.

In case of applying divergence theorem, we close the surface with a disk at $z = 0$.

$\small \nabla \cdot (x^4, y^4, z^4) = 4(x^3+y^3+z^3) = 4r^3(\cos^3\theta \sin^3\phi+ \sin^3\theta \sin^3\phi + \cos^3\phi)$

where $0 \leq r \leq R$.

So volume integral is,

$\displaystyle \small \int_0^{2\pi} \int_0^{\pi/2} \int_0^R 4r^5 \sin\phi (\cos^3\theta \sin^3\phi+ \sin^3\theta \sin^3\phi + \cos^3\phi) \ dr \ d\phi \ d\theta = \frac{\pi R^6}{3}$.

Now note the flux through the bottom disk is zero so flux through spherical surface is $\frac{\pi R^6}{3}$, the same answer we received through direct surface integral.

$\endgroup$
2
  • $\begingroup$ Ah yes thanks makes sense now. Guess the provided solutions were wrong. $\endgroup$
    – Chris
    Apr 21 at 19:47
  • $\begingroup$ You are welcome. Yes it is wrong or they mistyped the vector field / region. $\endgroup$
    – Math Lover
    Apr 21 at 19:50

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.