Let D be a "nice" region in the plane (where "'nice" means the boundary has a continuous parametrization and does not intersect itself). The boundary should be traversed in the counterclockwise direction. Suppose that is a vector field, and P and Q have continuous partial derivatives. Green's theorem relates the line integral around the boundary to the double integral over D:
Green's theorem is a special case of Stokes' theorem; to peek ahead a bit, is just the z component of the of , where is regarded as a 3-dimensional vector field with zero z component:
Example. Let D be the unit disk . Its boundary is the unit circle , which has the parametrization
Verify that Green's theorem holds for the line integral
First, I'll compute the line integral directly.
So
Apply the double angle formulas:
So I have
On the other hand, Green's theorem says the line integral is equal to
Convert to polar. The region is
So the integral becomes
The results agree.
Example. Verify Green's theorem for and the curve , .
The curve is the unit circle again, and the region D it encloses is the disk .
So the line integral is
Green's theorem says that the line integral is equal to the double integral
The results agree.
Example. Let R be the region bounded below by the x-axis, bounded on the right by for , and bounded on the left by for . Compute
The region is
By Green's theorem,
Example. Let D be the region bounded by , from to , and the x-axis. Compute
Assume that the boundary is traversed counterclockwise.
The region is
By Green's theorem,
If you compute the line integral directly, you need to parametrize the segment which makes up the base of the region and the curve. However, the curve is as x goes from to 0, because the boundary of the region is traversed counterclockwise.
Example. Let P be the parallelogram with vertices , , , and . Compute
Assume the boundary is traversed counterclockwise.
To do the line integral directly, I'd need to parametrize each side and compute the integral over each side. Rather than do four integrals, I'll use Green's theorem:
I don't need to compute the double integral; the area of a parallelogram is the length of the cross product of the vectors for two adjacent sides. and , so
Now , so the integral is .
In the last two examples, the double integral reduced to a number times the area of the region. You can use Green's theorem to find the area of a region D as follows.
Example. The trisectrix of MacLaurin is given by the parametric equations
(The trisectrix is the pedal curve of a parabola; the pedal point is the reflection of the focus across the directrix.)
The entire curve is traced out from to . The loop is traced out from to . Find the area of the region enclosed by the loop.
The area is
(You can integrate the second and third terms using the double angle formulas for sine and cosine.)
Example. (a) Parametrize . (Hint: Let .)
(b) Find the area enclosed by the loop of the curve.
(a) Following the hint, set . Then
Hence, .
is a line through the origin with slope t. Thus, the curve is being parametrized by the slope of the line joining the origin to a point on the curve.
(b) The loop is traced as t goes from 0 to .
Hence,
The area is
You can do this integral by letting , so . As t goes from 0 to , u goes from 1 to . The integral becomes
Here's how to do the antiderivative:
(Multiply out the u stuff on top, then divide each term by .)
Copyright 2018 by Bruce Ikenaga