Jacobian determinant

**Dr Zoidburg** · May 6th 2010, 08:15 PM

I'm going to rush through most of this stuff, as I feel I've got it correct. It's just the answer to the last question that's bugging me.
Question

Let R be the region in the xy-plane enclosed by the lines:

x + y = 0, x + y = 1, x – y = 1, x – y = 4.

(a) Draw a picture of the region R.
answer:
y= -x, y = 1-x, y=x-1 & y = x-4. It makes a rectangle with the corners (0.5,-0.5), (1,0), (2,-2), & (2.5,-1.5).

(b) Define a change-of-variables transformation by the equations

u = x + y & v = x – y.
What is the image S in the uv-plane of R under this transformation?
answer:
solving for x and y:
x=1/2(u+v) & y=1/2(u-v)
Putting these into the above equations I get the lines u=0, u=1, v=1/2 and v=2. It makes a rectangle with the corners (0,1/2), (0,2), (1,1/2) & (1,2)

(c) Solve the transformation equations for x and y in terms of u and v, and hence find the determinant of the Jacobian matrix.
answer:
$\tfrac{\partial x}{\partial u} = \tfrac{1}{2}, \tfrac{\partial x}{\partial v} = \tfrac{1}{2}, \tfrac{\partial y}{\partial u} = \tfrac{1}{2}, \tfrac{\partial y}{\partial v} = \tfrac{\text{-}1}{2}$

$\frac{\partial(x,y)}{\partial(u,v)} =\begin{vmatrix}\frac{1}{2} & \frac{1}{2}\\ \frac{1}{2} & \frac{\text{-}1}{2}\end{vmatrix}=\frac{\text{-}1}{2}$

(d)Use this change of variables to evaluate
$\int \int_R (x - y)e^{x^2-y^2}dA$
answer:
simplifying
$x-y = \tfrac{1}{2}(u+v)-\tfrac{1}{2}(u-v) = v$
$x^2-y^2 = (\tfrac{1}{2}(u+v))^2 - \tfrac{1}{2}(u-v))^2 = \frac{uv}{2}$
makes the equation:

$\int_{\frac{1}{2}}^{2}\int_{0}^{1}v.e^{\frac{vu}{2 }}(\tfrac{\text{-}1}{2}). \partial u\partial v$

inner integral:

$\int_{\frac{1}{2}}^{2}\int_{0}^{1}v.e^{\frac{vu}{2 }}(\tfrac{\text{-}1}{2}). \partial u\partial v = -e^{\frac{uv}{2}}\bigg|_{0}^{1} = \int_{\frac{1}{2}}^{2}\text{-}e^{\frac{v}{2}}+1. \partial v$

outer integral:
$\int_{\frac{1}{2}}^{2}\text{-}e^{\frac{v}{2}}+1. \partial v = v-2e^{\frac{v}{2}}\bigg|_{\frac{1}{2}}^{2} = (2 - 2e^{1})-(\tfrac{1}{2} - 2e^{\frac{1}{4}})= 1\tfrac{1}{2} - 2(e^{1}-e^{\frac{1}{4}})$

So is my answer to (d) correct? It's seems so unwieldy and unpleasant an answer considering how nice and easy the rest of the Question was. If anyone could check my work and affirm/negate my answers, please do!
cheers

**matheagle** · May 6th 2010, 10:27 PM

1 you need not find the inverse transformation

I would obtain ${\partial(u,v)\over\partial(x,y)}=-2$ instead, so our jacobian is absolute
value of the reciprocal of -2, i.e., +.5.

2 clearly $x^2-y^2=(x+y)(x-y)=uv$ not uv/2

3 you need the absolute value around the jacobian, so it's plus 1/2, not -1/2.

4 The region is 0<u<1 and 1<v<4.

$.5\int_1^4\int_0^1 ve^{uv}dudv =.5\int_1^4(e^v-1)dv$

**Dr Zoidburg** · May 7th 2010, 04:11 AM

thanks for that. Too many stupid small mistakes on my part add up to a nasty final answer!
One thing: why do you have the limits for v: 1<v<4 and not 1/2<v<2?

**matheagle** · May 7th 2010, 06:07 AM

just insert u and v into....

x + y = 0, x + y = 1, x – y = 1, x – y = 4.

you need not even draw this, it's too simple
somehow you have that 1/2 you incorrectly obtained via bad algebra
check that inverse transformation, I bet it's wrong
YOU don't need to obtain the inverse transform.
I get the upside down jacobian and take the reciprocal
Lots of books don't mention that you can do that
Plus you do need the absolute values too.
Is this from Stewart by the way?

Thread: Jacobian determinant

LinkBack

Thread Tools

Display

Jacobian determinant

Similar Math Help Forum Discussions

Jacobian determinant question

Derivation of Jacobian Determinant

The problem of Jacobian determinant

jacobian

Jacobian

Search Tags