I'm working on a problem where $F$ is a vector field $F=\langle x,y,z\rangle $ where the $E$ is the unit ball $ x^2+y^2+z^2 \le 1 $.

This is an easy surface integral to calculate using Divergence Theorem:

$$ \iiint_E {\rm div}(F)\ dV = \iint_{S=\partial E} \vec{F}\cdot d{\bf S}$$

But, to confirm the divergence theorem by calculating the surface integral directly, how would you set up the bounds on the double integral for a unit ball?

By the way, $div(\vec{F}) = 0$ in this case, and then it's a triple integral $\iiint_E 1 dV$. Since E is a solid ball and $1$ with in triple integral results in a volume calculation, the answer is $\frac{4\pi r^3}{3} = \frac{4\pi 1^3}{3} = \frac{4\pi }{3}$. So that's the final numerical answer I'm looking for when doing the surface integral in order to verify the divergence theorem.

  • $\begingroup$ ${\bf x}(x,y)=(x,y,\sqrt{1-x^2-y^2}), \ {\bf y}(x,y)=(x,y,-\sqrt{1-x^2-y^2})$ where $x^2+ y^2 \leq 1$. How about calculate double integral twice ? but in this case since $F=(x,y,z)$, we can calculate double integral easily. $\endgroup$ – HK Lee May 8 '14 at 5:14
  • $\begingroup$ Alternatively, in spherical coordinates: $r(\theta,\phi)=(\cos\theta\sin\phi,\sin\theta\sin\phi,\cos\phi)$, $0\leq\theta\leq 2\pi$,$0\leq\phi\leq\pi$ $\endgroup$ – A Nonny May 8 '14 at 5:33
  • 1
    $\begingroup$ I'm not really following why there are two parametrization here which look exactly the same. Perhaps I should be calculating the surface integral for a unit sphere? $\endgroup$ – Bob Shannon May 8 '14 at 5:34
  • 1
    $\begingroup$ If we cut a sphere along equator, we have two surfaces. So we need two : I edit the second parametrization ${\bf y}(x,y)=(y,x,-\sqrt{1-x^2-y^2})$ $\endgroup$ – HK Lee May 8 '14 at 5:56
  • $\begingroup$ Oh ok, I see now. Spherical coordinates may come in handy here then to prevent having to do two surface integrals. $\endgroup$ – Bob Shannon May 8 '14 at 5:59

After reading a few examples on the web and in my textbook I think I may be have figured this out. It seems that the hardest (or most tedious) part of this problem is actually performing the calculus. Indeed, I needed to calculate the surface integral of a unit sphere floating around in the given vector field, because a unit sphere is the surface that lines a unit ball that was given in the problem.

A unit sphere can be parametrized using spherical coordinates:

$$ r(\phi,\theta) = \sin(\phi)\cos(\theta)\hat{i} + \sin(\phi)\sin(\theta)\hat{j}+\cos{\phi}\hat{k}$$

By definition, a surface integral: $$ \iint_S \vec{F} \cdot d\vec{S}= \iint_D\vec{F}(\vec{r}(\phi,\theta))\cdot(\vec{r}_u \times \vec{r}_v) dA$$

If you have some time and decide to crank this out and set up the integral:

$$\vec{F}(\vec{r}(\phi,\theta)) = \cos(\phi)\hat{i}+\sin(\phi)\sin(\theta)\hat{j}+\sin (\phi)\cos(\theta)\hat{k} $$ $$ \vec{r}_\phi \times \vec{r}_\theta = \sin^2(\phi)\cos(\theta)\hat{i}+\sin^2(\phi)\sin(\theta)\hat{j}+\sin(\phi)cos(\phi)\hat{k}$$

Then: $$\vec{F}\cdot(\vec{r}_u \times \vec{r}_v) = \cos(\phi)sin^2(\phi)\cos(\theta)+\sin^2(\phi)sin^2(\theta)+\sin^2(\phi)\cos(\phi)\cos(\theta)$$

The integral is now ready to put in the oven:

$$\iint_D [\cos(\phi)sin^2(\phi)\cos(\theta)+\sin^2(\phi)sin^2(\theta)+\sin^2(\phi)\cos(\phi)\cos(\theta)] dA$$

$$= \int_0^{2\pi}\int_0^\pi [\cos(\phi)sin^2(\phi)\cos(\theta)+\sin^2(\phi)sin^2(\theta)+\sin^2(\phi)\cos(\phi)\cos(\theta)] d\phi d\theta$$

$$= 2 \int_0^\pi\sin^2(\phi)\cos(\phi) d\phi \int_0^{2\pi}\cos(\theta)d\theta\int_0^\pi\sin^3(\phi)d\phi\int_0^{2\pi}\sin^2\theta d\theta$$

$$= 0 + \int_0^\pi \sin^3(\phi)d\phi \int_0^{2\pi} \sin^2(\theta)d\theta $$ $$ = \frac{4\pi}{3} $$

  • 1
    $\begingroup$ Another way : Over unit sphere, normal vector is just ${\bf F}$ So $\iint_S {\bf F}\cdot d{\bf S} = \iint_S dS = 4\pi$ $\endgroup$ – HK Lee May 8 '14 at 6:39

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.