1

1.1

Multiple Integrals and Vector Calculus

Double integrals

Let f (x, y) be a real valued function dened over a domain IR2 . To start with, let us
assume that be the rectangle R = (a, b) (c, d). We partition the rectangle with node
points (xk , yk ), where
a = x1 < x2 < ...., xn = b, and c = y1 < y2 < ...., yn = d.
Let Rnm be the small sub-rectangle with above vertices. Now we can dene Upper and
lower Riemann sum as

U (Pn , f ) =
sup f (x, y)|Rnm |
n,m Rnm

and
L(Pn , f ) =


n,m

inf f (x, y)|Rnm |

Rnm

where |Rnm | is the area of the rectangle Rnm . Here we may dene the norm of partition Pn

as Pn  = max |xi xi1 |2 + |yi yi1 |2 . Then by our understanding of denite integral
i

we can dene the upper, lower integrals and f (x, y) is integrable if and only if
inf{U (P, f ) : P } = sup{L(P, f ) : P }.
The denite integral is dened as

f (x, y)dA = inf{U (P, f ) : P } = sup{L(P, f ) : P }.


It can be shown that

f (x, y)dA = lim S(Pn , f ), where S(Pn , f ) =

Pn 0

f (xk , yk )|Rnm |.

n,m

We have the following Fubinis theorem for rectangle:

Suppose f (x, y) is integrable over R = (a, b) (c, d), then



 b  d
 d  b
f (x, y)dA =
f (x, y)dy dx =
f (x, y)dx dy
R

In case of f (x, y) 0 we may interpret this as the volume of the solid formed by the
surface z = f (x, y) over the rectangle R. This is precisely the sum of areas of the cross
d
section A(x) = c f (x, y)dy between x = a and x = b. Since x varies over all of (a, b), this
b
sum is nothing but the integral a A(x)dx.

For any general bounded domain , we can divide the domain into small sub domains
k and consider the upper, lower sum exactly as above by replacing Rnm by k . Then
a function f (x, y) : IR is integrable if the supremum of lower sums and inmum of
upper sums exist and are equal. We may dene

f (x, y)dA = lim

Pn 0

f (xk , yk )|k |

The basic properties of the denite integral like integrability of f g, kf and domain
decomposition theorems holds in this case also.
We call a domain as y-regular if = {(x, y) : a x b, g1 (x) y g2 (x)} for some
continuous functions g1 , g2 . Similarly, is x-regular if = {(x, y) : c y d, h1 (y)
x h2 (y)}. A domain is called regular if it is either x-regular or y-regular. Then we have
the following Fubinis theorem for regular domains.
Theorem 1.1 Let f (x, y) be continuous over .
1. If is y-regular, i.e., = {(x, y) : a x b, g1 (x) y g2 (x)}. Then
 b 



g2 (x)

f (x, y)dA =

f (x, y)dy dx.

g1 (x)

2. If is x-regular, i.e., = {(x, y) : c y d, h1 (y) x h2 (y)}. Then







h2 (y)

f (x, y)dA =
R

f (x, y)dx dy.

h1 (y)

This theorem basically says that if a function is integrable over a domain , then the value
of integral is does not depend on order of integration. That is we can integrate with respect
to x rst followed by y or vice versa.
(x + y + xy)dA where is the triangle bounded by

Example 1: Evaluate the integral

y = 0, x = 1 and y = x.

Solution: The triangle is regular in both x and y. The given triangle is

{(x, y) : 0 x 1, 0 y x} = {(x, y) : 0 y x, y x 1}
2

Therefore, taking it as y regular we see that the domain is bounded below by y = g1 (x) =
0 and above by y = g2 (x) = x over x = 0 to x = 1. Hence,

 x=1  x

(x + y + xy)dA =
(x + y + xy)dy dx

x=0
1

(x2 +

=
0

y=0
2

x3
15
x
+ )dx =
2
2
24

Similarly, taking it as x regular, we see that the domain is bounded below y x = h1 (y) = y
and above by x = h2 (y) = 1 over y = 0 to 1. Hence

 1  1

15
(x + y + xy)dA =
(x + y + xy)dx dy =
24

y=0
x=y

Example 2: Evaluate the integral
(2 + 4x)dA where is the domain bounded by

y = x and y = x2 .
Solution:





(2 + 4x)dA =

(2 + 4x)dy dx
y=x2

x=0
 1

=
0

(2x + 2x2 4x3 )dx = 2/3



On the other hand, this is also equal to


(2 + 4x)dx dx.

y=0

x=y

Remark 1.1
1. Whenf (x, y) = 1, then we approximate the area of as A k k =

k f (xk , yk )|k | where f = 1. By the denition of Riemann integral this sum con
verges to dA as Pn  0.

2. As discussed in the beginning, when f (x, y) 0, the f (x, y)dA is the volume of
the solid bounded above by z = f (x, y) and below by .
Example: Find the area bounded by y = 2x2 and y 2 = 4x.
Solution: The two parabolas intersect at (0, 0) and (1, 1). Hence the area is


2 x

dydx =

A=
0

2x2

2
(2 x 2x2 )dx = .
3

Example: Find the volume of the solid under the paraboloid z = x2 +y 2 over the bounded
domain R bounded by y = x, x = 0 and x + y = 2.
3

Solution: The domain of integration R is Y -regular bounded above by x + y = 2 and

below by y = x with x varying over (0, 1).

 1  y=2x

2
2
2
2
(x + y )dA =
(x + y )dy dx
V =
R

y=x
1

=
0

y
+ yx2 y=2x
y=x dx =
3

1
0

1
( ((2 x)3 x3 ) + x2 (2 2x))dx
3

Example: Find the volume of the solid bounded above by the surface z = x2 and below
by the plane region R bounded by the parabola y = 2 x2 , y = x.
Solution: The points of intersection of y = x, y = 2 x2 are x = 2, 1. So R = {(x, y) :
2 x 1, x y 2 x2 }. Therefore,


2x2

x dydx =

V =
x=2

y=x

x2 (2 x2 x)dx

Change of order

Consider the evaluation of integral R sinx x dA over the triangle formed by y = 0, x = 1 and
y = x. Since the can be extended as continuous function over R, by the basic properties of
Riemann integral the function is integrable. Now by Fubinis theorem, the value of integral
does not depend on the order of integration. As we noted earlier R is regular in x and y.
If we take it as x regular, then R = {(x, y) : 0 y 1, y x 1} and try to evaluate the
integral, then

 1  1

sin x
sin x
dA =
dx dy.
x
R x
0
x=y
This is singular integral and dicult to evaluate.
But when we consider R to be y-regular, we see that R = {(x, y) : 0 x 1, 0 y x}.
Then the given integral is
 1 x
 1

sin x
sin x
dA =
dydx =
sin xdx = 1 cos 1
R x
0
y=0 x
0
At times this technique can be used to evaluate some complicated denite integrals, for
example,
 ax bx
Example: Evaluate the integral 0 e e
dx, a, b > 0.
x
Solution: This integral is equivalent to

  b
 ax
e
ebx
xy
dx =
e dy dx
x
0
0
a
4

The domain of integration is the innite strip {(x, y) : 0 x , a y b}. Changing

the order of integration, we get


  b
 b 
xy
xy
e dy dx =
e dx dy
0

y=a

= ln

b
a

Double integrals in Polar form

Suppose we are given a bounded region whose boundaries are given by polar equations,
say f1 (r, ) = 0, f2 (r, ) = 0. Then we divide the region into smaller polar rectangles
whose sides have constant r, values.
Suppose f (r, ) is dened over a region R dened using the polar equations, R :
, g1 () r g2 (). Then we divide the r range by r, 2r, ...., mr and , +
, ..... + m = . Let A be the polar rectangle with sides rk r/2, rk + r/2 and
+ k, + (k + 1). Then we dene the Riemann sum as

f (rk , k )Ak .
Sn =
k

The area of small polar rectangle Ak is

Ak = area of outer sector area of inner sector = rk r.
As Pn  0,we get
Sn =


f (rk , )rk r

f (r, )rdrd.
R

Example: Find the area common to the cardioids r = 1 + cos and r = 1 cos .
Solution: Since the region is symmetric with respect to x -axis and y-axis, the required
area is
 /2  1cos
rdrd
A =4
=0
r=0
 /2
 /2
1

2
=4
(1 2 cos + cos )d = 2( 2 +
cos2 d)
2
2
0
0

Example: Evaluate
3ydA where R is the region bounded below by x-axis and above
R

by the cardioid r = 1 cos .

Solution: The given integral is equivalent to

  1cos
3ydA =
r sin rdrd.
R

=0

r=0

Example: Evaluate I =

ex dx.

Solution: Recall that 2I = ( 12 ). Using the Fubinis theorem, we may write


 
  
2
2
2
x2
y 2
I =
e dx
e dy =
e(x +y ) dA
0

Where the integration is over the rst quadrant (0, ) (0, ). So representing this in
polar form we integrate over {(r, ) : 0 r < , 0 2 }. Therefore, the above integral
becomes
 /2 

2
er rdrd = .
4
0
0

1.2

Triple (Volume) integrals

Let f (x, y, z) be a real valued function dened over a closed and bounded region of space
IR3 . For example the solid ball or rectangular box. Now we want to dene the denite
integral of f (x, y, z) over such regions.
We partition the region by small planes parallel to the coordinate axes. Then we obtain
small rectangular cubes over which the function will be approximated by f (xk yk , zk ). We
form the Riemann sum

f (xk , yk , zk )|k |,
Sn =
k

where |k | is the volume of the small rectangle. Now by our understanding of Riemann
sums we choose renement of partitions in such way that maxk |k | 0. Then we obtain
the denite integral as


f (x, y, z)dV = lim Sn .

n

Evaluation of integrals in three dimensions is done again using Fubinis theorem. In

this case again Fubinis theorem states
Theorem 1.2 Suppose f (x, y, z) is integrable over IR3 , then
  
  

f (x, y, z)dV =
f (x, y, z) dz dy dx =
f (x, y, z) dx dz dy

x y z
x z y
  
  
f (x, y, z) dy dx dz =
f (x, y, z) dx dy dz
z x y
z y x
  
  
=
f (x, y, z) dz dx dy =
f (x, y, z) dy dz dx
y

To evaluate the triple integrals we follow the following steps:

6

1. Draw a line parallel to z axis that passes through the point (x, y) of R where R is
the projection of onto IR2 .
2. Identify the upper surface and lower surface through which this line passes at most
once.
3. Identify the upper curve and lower curve of the projection R and limits of integration.
It is easy to see from the denition, the volume of is



|k | = lim
1 |k | =
1 dV
V = lim
k

Example: Find the volume of the region bounded by the surfaces z = x2 + 3y 2 and
z = 8 x2 y 2 .

Solution: The volume is V =
dzdydx, where is bounded above by the surface

2
2
z = 8 x y and below by the surface z = x2 + 3y 2 . Therefore, the limits of z are from
z = x2 + 3y 2 to z = 8 x2 y 2 .
The Projection of on xy-plane is the solution of
8 x2 y 2 = x2 + 3y 2

= x2 + 2y 2 = 4.

Therefore the limits of x and y are to be determined by R : x2 + 2y 2 = 4. Hence

   8x2 y2
V =
dzdA
R y=x2 +3y 2
 
2
2

(4x )/2

=
2

(4x2 )/2

(8 2x2 4y 2 )dydx
(4x2 )/2

4
(8 x2 )y y 3

3
2
y= (4x2 )/2


4 2 2
=
(4 x2 )3/2 dx = 8 2.
3 2

Example: Find the volume of the region bounded by x + z = 1, y + 2z = 2 in the rst

quadrant.
Solution: Draw line parallel to z-axis and note that the upper surfaces are: 2z + y = 2
over triangle bounded by x = 0, y = 1y = 2x and z = 1 x over the triangle bounded by
y = 0, x = 1, y = 2x. Therefore,
 2  y/2  2y
 1  2x  1x
2
V =
dz dx dy +
dz dy dx
y=0

x=0

z=0

x=0

y=0

z=0

On the other hand, by rst drawing the line parallel to x-axis, we get


22z

1z

dx dy dz

V =
z=0

y=0

x=0

Taking the line parallel to y-axis we get



1x

22z

dy dz dx

V =
x=0

z=0

y=0

2 cos(x2 )

dx dy dz.
z
z=0 y=0 x=2y
Solution: Note that the projection of onto xyplane is the triangle bounded by y =
0, x = 2 and x = 2y. So changing the order of integration in x and y, we get
Example: (Order of integration) Evaluate

z=0
4

x=0
2

x/2

I=
=
z=0

x=0

y=0

2 cos(x2 )

dy dx dz.
z

x cos(x2 )

dx dz = 2 sin 4.
z

Substitutions in multiple integrals

Suppose a domain G in uv-plane is transformed onto a domain of xy-plane by a transformation x = g(u, v), y = h(u, v). Then any function of x, y may be written as a function
of u, v. Then the relation between the double integral over G and is


f (x, y)dx dy =
f (g(u, v), h(u, v))|J(u, v)|du dv

where J is the Jacobian given by




(x, y)  xu xv 
J=
=

(u, v)  yu yv 


1+ y2

2x y
dxdy.
2
y/2
0
Solution: The domain of integration is a parallelogram with vertices (0, 0), (1, 0), (3, 4) and
(2, 4). One has to divide the domain into 3 domains. Instead we can take the transformation
, v = y2 . Then the inverse transformation is x = u + v, y = 2v. Then
u = 2xy
2
Example: Evaluate the integral I =


 1 1

J =
 0 2
8




 = 2.


Under this transformation, the parallelogram is transformed into cube with vertices (0, 0), (1, 0)(1, 2)
and (0, 2). Now by change of variable formula

 2 1
I=
f (u + v, 2v)2dudv =
2udu dv = 2.
0

1x

Example: Evaluate the integral I =

0

x + y(y 2x)2 dA.

Solution: The given domain is the triangle bounded by x = 0, y = 0 and x + y = 1.

In this case the integrand is complicated....so we can take transformation u = x + y and
v = y 2x. Under this transformation, the given triangle will be transformed into triangle
bounded by v = u, v = 2u and u = 1. The inverse of this transformation is x = uv
and
3
2u+v
y = 3 . Hence the Jacobian


 1/3 1/3 


J =
 = 1/3.
 2/3 1/3 
Hence

I=
0



uv 2 dv du

v=2u

2
2
Example: Evaluate the integral I = R (2xdA
2 y 2 )2 over R : x + y 1.
Solution: Taking the transformation x = r cos , y = r sin , we get


 cos r sin 


J =
 = r.
 sin r cos 

By substitution formula,


I=
0

r=0

r dr d
= 2
(2 r2 )2

2
1

dt
= /2.
2t2

Substitution formula for triple integrals

As discussed above suppose a three dimensional domain G is transformed onto a domain
D with a transformation x = x(u, v, w), y = y(u, v, w), z = z(u, v, w), then


f (x, y, z)dV =
F (u, v, w)|J(u, v, w)|dV
D

where



 xu xv xw 


(x, y, z)


J=
=  yu yv yw  .

(u, v, w) 
 zu zv zw 
9

The main idea of the proof is as follows. Let (u, v), (u + u, v), (u + u, v + v) and
(u, v + v) be the vertices of the rectangle in the uvplane. Let Ak be its area element.
Under the transformation this points are mapped to (x1 , y1 ) = (g(u, v), h(u, v)), (x2 , y2 ) =
(g(u + u, v), h(u + u, v)), (x3 , y3 ) = (g(u + u, v + v) and (x4 , y4 ) = (g(u, v +
v), h(u, v + v)). Then by Taylors theorem
g(u + u, v) = g(u, v) +

g
u + o((u)2 )
u

g
g
u +
v + o((u)2 ) + o((v)2 )
u
v
Then the area of the rectangle in xy-plane Ak is
g(u + u, v + v) = g(u, v) +

Ak |(x3 x1 )(y3 y1 ) (x3 x2 )(y3 y2 )|

|J|uv + o((u)2 ) + o((v)2 )
Taking this as the area in the Riemann sum of f (x, y) we get the required formula.

Example: Evaluate

(x2 y + 3xyz)dV where R = {(x, y, z) : 1 x 2, 0 xy

2, 0 z 1}.
solution: We take the transformation u = x, v = xy and w = z. Then the planes x = 1, 2
transforms to u = 1, 2. The plane y = 0 transforms to v = 0. The surface xy = 2 transforms
to v = 2. Then the Jacobian J is


 1 0 0 

1 

=  y x 0  = x = u.


J
 0 0 1 
Now by substitution formula,


1
(uv + 3vw) dw dv du
u
v=0 w=0
u=1
2 2
3v
(v + )dv du
=
2u
1 2 0
3
(2 + )du = 2 + 3 ln 2.
=
u
1

I=

Cylindrical coordinates: A point P in the space (IR3 ) is represented by (r, , z) where

r, are polar coordinates of the projection of P on to xy-plane and z is the z distance of
10

the projection from P . When

the Jacobian is

 xr


J =  yr

 zr

we take the transformation x = r cos , y = r sin , z = z,

 

x xz   cos r sin 0 
 

y yz  =  sin r cos 0  = r
 

z zz   0
0
1 

Example: Find the volume of the cylinder x2 + (y 1)2 = 1 bounded by z = x2 + y 2 and

z = 0.
Solution: Drawing a line parallel to z axis, we see that the limits of z are from 0 to x2 + y 2
and the projection onto xy-plane is the disc:R : x2 + (y 1)2 1. Therefore,
   x2 +y2
V =
dzdA
R

z=0

Now taking the cylindrical coordinates x = r cos , y = r sin , z = z we get the projection
to be
r2 cos2 + r2 sin2 2r sin = 0
i.e., r(r 2 sin ) = 0 = r = 0 to r = 2 sin


2 sin

r2

V =

rdz dr d
=0


r=0
 2 sin

=
=4

=0 r=0

4

z=0

r3 dr d

sin d =

5
4

Spherical polar coordinates: A point P in the space is represented by (, , ) where

is the distance of P from the origin, is the angle made by the ray OP with positive z
axis and is the angle made by the projection of P (onto xy-plane) with positive x-axis.
So it is not dicult to see that the relation with cartesian coordinates: The projection of
P on xy-plane has polar representation: x = r cos , y = r sin where r is the distance of
the projected point to origin. Therefore r = sin . From the denition of it is easy to
see that z = cos and
x = sin cos , y = sin sin , z = cos
The Jacobian in

 x


J =  y

 z

this case is
 
x x   sin cos sin sin cos cos
 
0
y y  =  sin sin r cos sin
 


cos
0
sin
z z

To nd limits of integration in spherical coordinates,

11





 = 2 sin



1. Draw a ray from the origin to nd the surfaces = g(, ), = g2 (, ) where it

enters the region and leaves the region.
2. Rotate this ray away and towards z-axis to nd the limits of
3. Identify the projection R of the domain on the xy-plane and polar form of R to write
the limits of .

dV

where is the unit ball x2 + y 2 + z 2 1.
Example: Evaluate
2
2
2
1+x +y +z

Solution: Going to spherical polar coordinates x = sin cos , y = sin sin , z =

cos , we get
 1  2 
2

sin d d d
I=
1 + 2
=0 =0 =0
 1

2
1

d = 4( 2 ln( 2 + 1)).
=(2 2)
2
1 + 2
0

Example: Evaluate I =

xdV where is the part of the ball x2 + y 2 + z 2 4 in the

rst octant.
Solution: Going to cylindrical coordinates, x = sin cos , y = sin sin , z = cos .
Since the domain is the ball of radius 4, we see that the limits of are from 0 to 2. Again
since it is cut by the xy-plane below, varies from 0 to /2. The projection is the circle
in the rst quadrant with radius 2. So varies from 0 to /2. Hence,


/2

/2

I=

sin cos (2 sin ) d d d

=0 =0
=0
/2  /2

sin2 cos d d =

=4
=0

=0

Surface Area and Surface integrals

Consider a surface S dened with f (x, y, z) = c. Let R be its projection on xy-plane.
Assume that this projection is one-one, onto. Let Rk be a small rectangle with area Ak
and let k be the piece of surface above this rectangle. Let Pk be the tangent plane
at (xk , yk , zk ) of the surface k . Now consider the parallelogram with Pk and Ak as
upper and lower planes of the parallelogram. We approximate the area of the surface with
the area of the tangent plane Pk .
Now let p be the unit normal to the plane containing Rk and f is the normal to the
surface. Let uk , vk be the vectors along the sides of the tangent plane Pk . Then the area
12

of Pk is |uk vk | and uk vk is the normal vector to Pk . Thus f and uk vk are

both normals to the tangent plane Pk .
The angle between the plane Ak and Pk is same as the angle between their normals.
i.e., the angle between p and uk vk . From the geometry, the area of the projection of this
tangent plane is |(uk vk ) p|(proof of this can be seen in Thomas calculus Appendix 8).
i.e.,
Ak = |(uk vk ) p| = |(uk vk )||
p|| cos(angle between (uk vk ) and p)
In other words,
Ak
| cos k |
where k = angle between (uk vk ) and p. This angle can be calculated easily by noting
that f and uk vk are both normals to the tangent plane.
(This formula is simple in case of straight lines: Let OP be the line from origin and let
R be the projection of P on x-axis. Then OR = OP cos where is the angle between
OP and OR. Now imagine the Area of plane is nothing but sum of lengths of lines.)
Pk | cos k | = Ak or Pk =

So
|f p| = |f ||
p|| cos k |
Therefore,
Surface Area

Pk =

This sum converges to

 |f |
Ak
|f p|
k



|f |
dA
|
R |f p
where R is the projection of S on to one of the planes and p is the unit normal to the plane
of projection.
Example: Find the surface area of the curved surface of paraboloid z = x2 + y 2 that is
cut by the plane z = 2.
Solution: The equation of surface is f (x, y, z) = z x2 y 2 = 0. Clearly this is one-one
from xy-plane to IR3 . So the projection of the surface {(x, y, x2 + y 2 ) : x2 + y 2 2} is the
Hence
disc R : x2 + y 2 2. Since the plane of projection is xy-plane, p = k.
Surface Area =

f = 2xi 2yj + k

|f |
dA
S=

R |f p
 
=
4x2 + 4y 2 + 1 dA
R

13

Going to polar coordinates x = r cos , y = r sin ,

 2  2
1 + 4r2 r dr d = 13
S=
0

r=0

Example: Find the surface of the cap obtained by cutting the hemisphere x2 +y 2 +z 2 =

2 by the cone z = x2 + y 2 .
Solution: The equation of surface is f (x, y, z) = x2 + y 2 + z 2 2 = 0 and we can
The projection is obtained by solving
take the projection onto xy-plane. So p = k.

x2 + y 2 + z 2 = 2, z = x2 + y 2 . i.e., R = x2 + y 2 = 1.
f = 2xi + 2yj + 2z k

|f p| = 2z = 2 2 x2 y 2
Therefore, using polar coordinates x = r cos , y = r sin ,


2

dA
S=
2 x2 y 2
R
 2  1

=
r dr d = 2(2 2).
2 r2
0
0
Surface Integrals
Let g(x, y, z) be a function dened over a surface S. Then we can think of integration of
g over S. Suppose, a surface S is heated up, we have a temperature distributed over this
surface. Let T (x, y, z) be the temperature at (x, y, z) of the surface. Then we can calculate
the total temperature on S using the Riemann integration.
Let R be the projection of S on the plane. We partition R into small rectangles Ak .
Let Sk be the surface above the Ak . We approximate this surface area element with
its tangent plane Pk . As we rene the rectangular partition this PK approximated the
Sk . Then the total temperature may be approximated as



Ak
|f |
=
dA
g(xk , yk , zk )Pk =
g(xk , yk , zk )
g(xk , yk , zk )
| cos k |
|f p|
k
k
k
where p is the unit normal to R or the plane of projection. Now taking limit n , we
get


|f |
dA
g(x, y, z)dS =
g(x, y, z)
|f p|
S
R
If the surface is dened as f = z h(x, y) = 0, then


|f |
g(x, y, z)dS =
g(x, y, h(x, y))
dA.

|f k|
S
R
14

Example: Integrate g(x, y, z) = z over the surface S cut from the cylinder y 2 +z 2 = 1, z
0, by the planes x = 0 and x = 1.
Solution: f = y 2 + z 2 and this surface can be projected 1-1, onto to R of xy plane. This
projection is the rectangle with vertices (1, 1), (1, 1), (0, 1), (0, 1). So p = k

2 y2 + z2
1
|f |
=
=
|f p
|2z|
z
Therefore,




zdS =
S

z
R

1
dA = Area(R) = 2
z

Parametrizations of Surfaces: Let

r(u, v) = f (u, v)i + g(u, v)j + h(u, v)k
be a continuous vector function dened on a plane region R. The variable u and v are
parameters and R is the parameter domain. The range of r is called the surface S. We
assume that r is one-to-one on the interior of R so that S does not cross itself.
Examples:
1. A parametrization of the cone
z=


x2 + y 2 , 0 z 1

Here cylindrical coordinates provide everything we need. Let

x(r, ) = r cos , y(r, ) = r sin z =


x2 + y 2 = r

the domain of r, is 0 r 1 and 0 2. So the required parametrization is

0 r 1, 0 2.
r(r, ) = r cos i + r sin j + r k,
2. A parametrization of sphere
x 2 + y 2 + z 2 = a2
Here we take spherical coordinates. Let x = a sin cos , y = a sin sin , and z =
a cos , 0 , 0 2. Taking u = and v = , we get

r(, ) = a sin cos i + a sin sin j + a cos k,

15

0 , 0 2.

3. A parametrization of cylinder x2 + (y 3)2 = 9, 0 z 5

we take cylindrical coordinates,
x(r, ) = r cos , y(r, ) = r sin z = z
Substituting in x2 + (y 3)2 = 9 we get r2 + 6r cos = 0. Therefore, r = 6 sin .
0 , 0 z 5
r(, z) = 3 sin 2 i + 6 sin2 j + z k,
Denition: (Smooth surface): A parametrized surface r(u, v) is called smooth surface if
ru and rv are continuous and ru rv is never zero.
Surface Area: We approximate the surface area element by the parallelogram on the
tangent plane whose sides are determined by the vectors ru u and rv v. The total surface
area is approximately equal to the sum of area of of this parallelograms

S
|ru rv |uv
u

bb
This sum is a Riemann sum of the integral a a |ru rv | dudv. Therefore we have
au
The Surface area of smooth surface r(u, v) = f (u, v)i + g(u, v)j + h(u, v)k,
b, c v d is
 
b

|ru rv | dudv

S=
a

Problem: Find the surface are of the surface of the cone


z = x2 + y 2 , 0 z 1
Solution: We found a parametrization of the cone as
0 r 1, 0 2.
r(r, ) = r cos i + r sin j + r k,
We can nd that
|rr r |

r2 cos2 + r2 sin2 + r2 = 2r

Therefore,

Surface Area =
0

2rdrd = 2

Surface integrals: Let F (u, v) be a continuous function dened on the parametrized

surface S: r(u, v) : R S, where R : a u b, c v d. Then

 b d
F dS =
F (u, v)|ru rv |dudv
S

16


Problem: Evaluate the surface integral S (x + y + z)dS over the surface of cylinder
x2 + y 2 = 9, 0 z 4.
Solution: Using the cylindrical coordinates: x = 3 cos , y = 3 sin , z = z over the
parameter domain {(, z) : 0 2, 0 z 4}. The surface can be represented as

T (, z) = 3 cos i + 3 sin j + z k.

Then |T Tz | = 9 cos2 + 9 sin2 = 3. The given integral is equal to



(x + y + z)dS =
(3 cos + 3 sin + z)|T Tz |ddz
S
S
 4
 4  2
(3 cos + 3 sin + z)ddz = 6
zdz = 48.
z=0

1.3

=0

Line integrals and Greens theorem

In many physical phenomena, the integrals over paths through vector led plays important
role. For example, work done in moving an object along a path against a variable force
or to nd work done by a vector eld in moving an object along a path through the
eld. A vector eld on a domain in the plane or in the space is a vector valued function
f : IR3 IR3 with components say M, N and P , for example
F (x, y, z) = M (x, y, z)i + N (x, y, z)j + P (x, y, z)k
We assume that M, N, P are continuous functions. Suppose F represents a force through a t b is a smooth curve in
out a region in space and let r(t) = x(t)i + y(t)j + z(t)k,
the region. Then we introduce the partition a = t1 < t2 ..... < tn = b of [a, b].
If Fk denotes the value of F at the point on the corresponding to tk on the curve and Tk
denotes the curves unit tangent vector a this point. Then Fk Tk is the scalar component
of F in the director of T at tk . Then the work done by F along the curve is approximately
n


Fk Tk sk ,

k=1

where sk is the length of the curve between tk1 , tk . As the norm of the partition approaches zero, these sums approaches

 b
 b

dx
dy
F T ( )2 + ( )2 dt.
F T ds =
dt
dt
t=a
a
Now substituting T (t) =

r  (t)
,

|
r  (t)|

we get


F
r  (t)dt.

17

The following can be shown:

Theorem 1.3 The Line integral is independent of choice of parametrization.
Denition: The orientation of a parameterized curve is the direction determined by
increasing values of the parameter.
The line integral over a parametrized curve depends on the orientation. If we change the
orientation, then the integral is equal to 1 times: That is,


F dr =
F dr.
C

Denition: Conservative vector eld: A verctor eld F is called conservative vector eld
if the line integral depends only on the end points. Equivalently, the line integral over any
closed curve is zero.

That is if there exists a potential f : IR3 IR such that F = f .

In this case,
 Q

F d
r = f (Q) f (P ).
P

It is clear that if P = Q, then above integral is zero.

Example: Show that F is a conservative vector eld.

Example: Find the work done by F = 3x2i + (2xz y)j z k over the curve r(t) =
0 t 1 from origin to (1, 1, 1)
ti + t2j + t3 k,
. Therefore,
Solution: The tangent along the curve T is dr
dt


1
0


F T ds =

0 1

=
0

d
r
dt
dt

2
3t2 + t5 2t3 dt = .
3

Greens theorem in the plane

Let R be a closed bounded region in IR2 whose boundary C consists of nitely many

smooth curves. Let F (x, y) = F1 (x, y)i + F2 (x, y)j be continuous and has continuous
partial derivatives everywhere in some domain containing R. Then


 

F2 F1

dA =
F d
r,
x
y
R
C
where the line integral is along the boundary C of R such that R is on the left as we advance
on the boundary.
18

Example: Evaluate
Solution:


F d
r for F = (y 2 7y)i + (2xy + 2x)j and C : x2 + y 2 = 1.
C
 



F2 F1

dA = 9
dA = 9.
x
y
C
R
R
Remark: The region R is such that the boundary consists of smooth curves. The theorem
does not hold true for regions like punctured disc {(x, y) : x2 + y 2 1}\{0}. For example,
  F2 F1 

F1
x
F2
y
=
and R x y = 0. But the line
i+ 2
j. Then
take F = 2
2
2
x
y
x2+ y
 x +y

F d
r =
sin2 + cos2 = 2.
integral
C

F d
r =

Area of plane region:

Using Greens theorem, we can write area of a plane region as a line integral over the
boundary. Choose F1 = 0, F2 = x and then F= y, F2 = 0. This gives




dA =
xdy and
dA = ydx
R

respectively. The double integral is the area A of R. By addition we have

1
A = (xdy ydx)
2
2

Example: Area bounded by ellipse xa2 + yb2 = 1.

solution: Take x = a cos t, y = b sin t, 0 t 2. Then by above formula


1 2
1
A=
ab cos2 t (ab sin2 t) dt = ab
(xy  yx )dt =
2 0
2

1.4

Gauss and Stokes theorems

Let S be a smooth surface and we may choose unit normal n

at P of S. The direction
of n
is called positive normal direction of S at P . We call a smooth surface S orientable
surface if the positive normal at P can be continued in a unique and continuous way to
the entire surface. For example the Mobius strip is not orientable. A normal at a point P
of this strip is displaced continuously along a closed curve C, the resulting normal upon
returning to P is opposite to the original vector at P .
Gauss Divergence Theorem
Let be a closed, bounded region in IR3 whose boundary is a piecewise smooth orientable

surface S. Let F (x, y, z) be a continuous function that has continuous partial derivatives
in some domain containing . Then



F n
dS
divF dV =

19

where n
is the outer unit
  normal vector of S.

F .
ndA where is the boundary of the domain inside the
Example: Evaluate

cylinder x2 + y 2 = 1 and between the planes z = 0, z = x + 2 and F = (x2 + yez )i + (y 2 +

ze2 )j + (z 2 + xey )k.

Solution: With the given F , it is not dicult to obtain, F = 2x + 2y + 2z. By

Divergence theorem


F n
dS =







x+2

2(x + y + z)dV = 2

(x + y + z)dz dxdy
x2 +y 2 1

z=0

Stokess theorem
Let S be a piecewise smooth oriented surface with boundary and let boundary C be a simple

closed curve. Let F be a continuous function which has continuous partial derivatives in
a domain containing S. Then



F d
r
( F ) n
dS =
C

where n
is a unit normal vector of S and, depending on n
, the integration around C is
taken in the way that S lies in the left of C. Here n
is the direction of your head while
moving along the boundary with surface on your left.


r where F = x2 y 3i + j + z k and C The intersection of the

Example: Evaluate C F d
cylinder x2 + y 2 = 4 and the hemisphere x2 + y 2 + z 2 = 16, z 0.
Solution: The intersection of cylinder and sphere is the boundary of cylinder on the plane

The projection R of S on
z = 12. The unit normal to the surface is n
= 14 (xi + yj + z k).

|f |
= z4 . Hence by Stokes
the xy-plane is the disc x2 + y 2 2 , F = 3x2 y 2 k and |f

p|
theorem


3
4
( )x2 y 2 z dA
F dr =
4
z
C
R
 2  2
(r2 cos2 )(r2 sin2 )rdrd = 8.
=3
=0

r=0

Suppose S1 , S2 be two surfaces having the same boundary curve C. An important consequence of Stokes theorem is that ux through S1 or S2 is same.
Example: Suppose S is a surface of a light bulb over the unit disc x2 + y 2 = 1 oriented

2
2

with outward pointing normal. Suppose F = ez 2z xi + (sin(xyz) + y + 1)j + ez sin(z 2 )k.



Compute S ( F ) n
dS.
20

Solution: Enough to take any surface with boundary x2 + y 2 = 1. So we take the unit

disc x2 + y 2 1, z = 0. Then F on this is F = xi + (y + 1)j. Then F = 0. Hence



F d
r = 0.
C
References
1. Thomas Calculus, Chapters 15, 16
2. Advanced engineering Mathematics, E.Kreyszig, Chapter 9

21