2
$\begingroup$

I am following the discussion here: integration in five dimensions space

I am doing this problem: Consider the differential form

$$a=p_1 \, dq_1+p_2 \, dq_2-(p_1^2+p_2^2+q_1^2+q_2^2) \, dt\text{ in }\mathbb R^5=(p_1,p_2,q_1,q_2,t).$$

(a) Compute the differential $da$ and the form $a\wedge da$.

(b) Evaluate the integral $\int_S da\wedge da$ where $S$ is the 4-dim surface in $R^5$ defined by the equation $p_1^2+p_2^2+q_1^2+q_2^2=t$ and the inequality $1\le t\le 2$.

Here is the result from the previous discussion, and I still have some problems about the following work.

We showed that the integration can be evaluated by$$\int_{\partial S}a\wedge da$$(using Stokes's theorem), where $$\partial S:=A\cup B:=p_1^2+p_2^2+q_1^2+q_2^2=1\cup p_1^2+p_2^2+q_1^2+q_2^2=2$$ and $$a\wedge da=p_1dq_1\wedge dp_2\wedge dq_2+p_2dp_2\wedge dp_1\wedge dq_1+terms\quad with\quad dt.$$

So I just start to calculate the above integral(if $dp_1\wedge dp_2\wedge dq_1\wedge dq_2\wedge dt>0$): $$\int_{A\cup B}p_1dq_1\wedge dp_2\wedge dq_2+p_2dp_2\wedge dp_1\wedge dq_1$$

Firstly, I use the parametric form for $A$ (with radius 1): $p_1=\cos\theta_1$, $p_2=\sin\theta_1\cos\theta_2$, $q_1=\sin\theta_1\sin\theta_2\cos\theta_3$, $q_2=\sin\theta_1\sin\theta_2\sin\theta_3$, where $\theta_{1,2}\in[0,\pi], \theta_3\in[0,2\pi].$

Secondly, I calculate the Jacobi $$ \left. \begin{array}{c} p_1\\ p_2\\ q_1\\ q_2\\ \end{array} \right. \left( \begin{array}{ccc} -\sin\theta_1&0&0\\ \cos\theta_1\cos\theta_2&-\sin\theta_1\sin\theta_2&0\\ \cos\theta_1\sin\theta_2\cos\theta_3&\sin\theta_1\cos\theta_2\cos\theta_3&-\sin\theta_1\sin\theta_2\sin\theta_3\\ \cos\theta_1\sin\theta_2\sin\theta_3&\sin\theta_1\cos\theta_2\sin\theta_3&-\sin\theta_1\sin\theta_2\cos\theta_3\\ \end{array} \right)$$

Thirdly, I do the integration of the first part on $A$: $$\int_{A}-p_1dp_2\wedge dq_1\wedge dq_2.$$ This is equal to(plug in the parameters form):

$$\int_0^{\pi}\int_0^{\pi}\int_0^{2\pi}(-\cos\theta_1)\space \det(2,3,4\,\text{rows of Jacobi})\space d\theta_1d\theta_2d\theta_3.$$

Am I doing the right way to calculate the integral?(So that I could continue with the rest three terms)

Another question is how to determine the order of the parameters wedge product?Is it $d\theta_1d\theta_2d\theta_3$, or some other order? What determine this order? I need to plug the parameters into $dp_2\wedge dq_1\wedge dq_2$ to see the result whether it is $d\theta_1d\theta_2d\theta_3$ or not, right?

Thank you very much!

$\endgroup$
  • 5
    $\begingroup$ Goodness. This is painful. Why on Earth would someone assign this? This is an exercise in tedium if I've ever seen it... $\endgroup$ – Cameron Williams Apr 14 '14 at 23:41
  • $\begingroup$ This is qualify exam. I am not very familiar with this topic, so I am here to ask some some tricks or other small points in practice that I need to know. $\endgroup$ – breezeintopl Apr 15 '14 at 0:47
1
$\begingroup$

(a) is a simple calculation.

(b) is the difference of volumes of two 4-balls of given radii.

Let $H=p_1^2+p_2^2+q_1^2+q_2^2.$ Let $D$ be the subset of $\mathbb R^4$ given by $1\leq H\leq 2$. Then $S$ is the graph of the function $t=H$ over $D$. Now change variables in the integral to integrate over $D$ instead of $S$ and you get the volume of $D$ (since $da\wedge da$ pulls back to the volume form on $D$).

Let $c$ be the volume of the unit 4 ball. Then the volume of the 4 ball of radius $r$ is $cr^4$. So the required integral is $4c-c=3c$.

To calculate $c$ use induction. Consult any vector calculus textbook.

$\endgroup$
  • $\begingroup$ Thank you for your answer! So you do not use Stokes formula and calculate directly?(I actually don't understand pull back) what about the first two terms $p_1dq_1+p_2dq_2$? $\endgroup$ – breezeintopl Apr 16 '14 at 17:35
  • $\begingroup$ Pull back just means change of variable in a differential form. Just make the change of variable $t=H(p_1,...)$ and you get that $da\wedge da$ changes to the volume form in $\mathbb R^4$. $\endgroup$ – Gil Bor Apr 16 '14 at 18:12

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.