MULTIVARIABLE INTEGRAL CALCULUS

1. INTRODUCTION

• Revision: Consider a function f which is defined for a ≤ x ≤ b, we

b−a
divide the interval [a, b] into n intervals of equal width ∆x = n . Let
a = x 0 < x 1 < x 2 < ... < x n = b be the endpoints of these intervals and
x 1∗ , x 2∗ , ..., x n∗ are points such that x i∗ ∈ [x i−1, x i ]. The definite integral of
f from a to b is defined as
Z b n
X
f (x)d x = lim f (x i∗ )∆x
n→∞
a i=1

if the limit exists.

1
2

• Motivation: Consider a thin cardioid plate occupying some region D

with variable density σ(x, y). How to find the mass of the plate?

• Step 1: Divide the region D into small subregions using grid or mesh.

j *
(xij, yij)
*

1 2 i

• Step 2: In the (i, j) rectangle, choose a point (x i∗j , yi∗j ) inside this rec-
tangle. Assume that the density σ is constant on this rectangle with
value given by σ(x i∗j , yi∗j ), then the mass of this rectangle is ∆mi j =
σ(x i∗j , yi∗j )∆Ai j , where ∆Ai j is the area of this rectangle.
 3

• Step 3: Obtain an approximation of the mass of the thin plate by sum-

ming ∆mi j over all rectangles inside D:
X
M≃ σ(x i∗j , yi∗j )∆Ai j
rectangle (i, j) is inside D
X
= σ(x i∗j , yi∗j )∆x i ∆ y j
rectangle (i, j) is inside D

• Step 4: Reduce the size of the grids and repeat Step 2 and Step 3.

• Step 5: The mass of the thin plate is defined as the limit when the grid
size approaches zero, if the limit exists.
!
X
M = lim lim σ(x i∗j , yi∗j )∆x i ∆ y j
∆x→0 ∆ y→0
rectangle (i, j) is inside D
ZZ
:= σ(x, y)dA
D
4

2. MULTIPLE INTEGRALS OVER RECTANGULAR DOMAINS

• We first consider the case where

 2

D = R = [a, b] × [c, d] = (x, y) ∈ R : a ≤ x ≤ b, c ≤ y ≤ d

is a rectangle. We want to compute

ZZ
f (x, y)dA
R

where we can assume that f (x, y) is the density function, so that the
integral is the total mass M of the rectangle.

• Divide [a, b] into m-equal subintervals [x 0, x 1], [x 1, x 2 ], . . . , [x m−1 , x m ]

b−a
of length ∆x = each,
m

x i = a + i∆x, x 0 = a, x m = b;

and divide [c, d] into n-equal subintervals [ y0 , y1 ], . . . , [ yn−1 , yn ] of length

d−c
∆y = each,
n

y j = c + j∆ y, y0 = c, yn = d.

This division gives rise to m × n rectangles labeled by (i, j), 1 ≤ i ≤ m

and 1 ≤ j ≤ n, where

R i j = [x i−1, x i ] × [ y j−1 , y j ],
 5

has area ∆A = ∆x∆ y. Choose a point (x i∗j , yi∗j ) in R i j , we can then

approximate the mass of R i j by

f (x i∗j , yi∗j )∆A.

Adding up the contribution from all the rectangles, it follows that

m X
X n
M≈ f (x i∗j , yi∗j )∆A.
i=1 j=1

As m and n gets larger, we should expect this approximation is getting

better. Therefore
m X
X n
M = lim lim f (x i∗j , yi∗j )∆A
m→∞ n→∞
i=1 j=1

• Definition: The double integral of f (x, y) over a rectangle R = [a, b] ×

[c, d] is
ZZ m X
X n
f (x, y)dA = lim lim f (x i∗j , yi∗j )∆A
m→∞ n→∞
i=1 j=1
R

if the limit exists.

6

• Examples of multiple integrals:

(a) The total electrostatic charge on a region D carrying charge with
density ρ(x, y) is
ZZ
Q= ρ(x, y)dA
D

(b) The force exerted on a region D by a fluid exerting pressure p(x, y)

is
ZZ
F= p(x, y)dA
D

(c) The area of a region D is

ZZ
A= dA
D

(d) The volume under the surface z = f (x, y) above the region D in the
(x, y)-plane is
ZZ
V= f (x, y)dA
D
 7

(e) For a lamina with density σ(x, y) and occupying a region D, the
moment about the x-axis is
ZZ
Mx = yσ(x, y)dA
D

The moment about the y-axis is

ZZ
My = xσ(x, y)dA
D

The centre of mass (x̄, ȳ) is then given by

My Mx
x̄ = , ȳ =
M M
ZZ
where M = σ(x, y)dA is the mass of the lamina.
D

(f) For a lamina with density σ(x, y) and occupying a region D, the
moment of inertia about the x-axis is
ZZ
Ix = y 2σ(x, y)dA
D

The moment of inertia about the y-axis is

ZZ
Iy = x 2σ(x, y)dA
D
8

3. ITERATED INTEGRALS

• Let f (x, y) be a function defined over a rectangle [a, b] × [c, d]. Then
Zd
A(x) = f (x, y)d y is a function of x on [a, b]
Zc b
B( y) = f (x, y)d x is a function of y on [c, d].
a
We can define
Z b Z b –Z d
™ Z bZ d
A(x)d x = f (x, y)d y d x = f (x, y)d y d x
a a c a c

and
Z d Z d –Z b
™ Z dZ b
B( y)d y = f (x, y)d x d y = f (x, y)d x d y
c c a c a

respectively. They are called iterated integrals or repeated integrals.

 9

• Example: Evaluate the iterated integrals:

Z 4Z 5 Z5Z 4
(a) (x + y 2)d y d x (b) (x + y 2)d x d y
2 −1 −1 2

Solution:
Z 4Z 5 Z 4   3  y=5
y
(a) (x + y 2 )d y d x = x y +  dx
2 −1 2
3 y=−1
Z 4 • ˜ • ˜‹
125 1
= 5x + − −x − dx
2
3 3
Z4
= (6x + 42) d x
2
 x=4
2
= 3x + 42x  x=2

= 48 + 168 − (12 + 84) = 120

Z5Z4 Z5 2  x=4
2 x 
2
(b) (x + y )d x d y = +xy  dy
−1 2 −1
2 x=2
Z5
   

= 8 + 4 y2 − 2 + 2 y2 d y
−1
Z 5
2


= 6 + 2y dy
−1
3 5

2y 
= 6y +
3 −1
 ‹
250 2
= 30 + − −6 − = 120
3 3
10

• Fubini’s Theorem: If f (x, y) is continuous on the rectangle R = [a, b]×

[c, d], then
ZZ Z bZ d Z dZ b
f (x, y)dA = f (x, y)d y d x = f (x, y)d x d y.
a c c a
R

Idea of Proof:
ZZ X n
m X
f (x, y)dA = lim lim f (x i∗j , yi∗j )∆x∆ y
m→∞ n→∞
i=1 j=1
R

Choose x i∗ in the interval [x i−1, x i ] and choose y ∗j in the interval [ y j−1 , y j ].

Then (x i∗ , y ∗j ) ∈ [x i−1 , x i ] × [ y j−1 , y j ]. Therefore, we can choose [x i∗j , yi∗j ]
to be [x i∗ , y ∗j ]. Therefore,
ZZ m
X n
X
f (x, y)dA = lim lim f (x i∗ , y ∗j )∆ y∆x
m→∞ n→∞
i=1 j=1
R

By definition,
n
X Z d
lim f (x i∗ , y ∗j )∆ y = A(x i∗ ) = f (x i∗ , y)d y
n→∞
j=1 c

Therefore,
ZZ m
X Z b Z bZ d
f (x, y)dA = lim A(x i∗ )∆x = A(x)d x = f (x, y)d y d x.
m→∞
i=1 a a c
R
 11
ZZ
• Example: Evaluate the multiple integrals x cos(x y)dA, where R =
R
[0, π/2] × [−1, 2].

Solution:
π
ZZ Z 2
Z 2
x cos(x y)dA = x cos(x y)d y d x
0 −1
R
Z π
2
y=2
= sin(x y)| y=−1 d x
0
Z π
2
= (sin(2x) + sin x) d x
0
 x= π
cos(2x)  2
=− − cos x 
2 x=0
 ‹
cos π π cos 0
=− − cos − − − cos 0
2 2 2
=2
12

4. MULTIPLE INTEGRALS OVER GENERAL REGIONS

• Definition: A plane region D is said to be of Type I if it lies between

the graphs of two continuous functions of x, i.e.,

D = {(x, y) : a ≤ x ≤ b, g1(x) ≤ y ≤ g2 (x)} .

y y = g2(x)
x=a x=b

y = g (x)
1
0 x

FIGURE 1. Type I Region

• Definition: To evaluate the integral

ZZ
f (x, y)dA,
D

over a type I region D, choose c and d such that

c ≤ g1(x) ≤ g2(x) ≤ d
 13

for all a ≤ x ≤ b. Define F (x, y) on the rectangle R = [a, b] × [c, d]

such that 

 f (x, y), if (x, y) ∈ D
F (x, y) =

0, otherwise.
Then we define
ZZ ZZ
f (x, y)dA = F (x, y)dA
D R
Z bZ d
= F (x, y)d y d x
a c
Z bZ g2 (x)
= f (x, y)d y d x
a g1 (x)

y
y=a
x = h (y)
1

x =h2(y)

y=b
0 x

FIGURE 2. Type II Region

• Definition: A plane region D is said to be of Type II if it can be ex-

pressed as

D = {(x, y) : c ≤ y ≤ d, h1 ( y) ≤ x ≤ h2( y)} .

14

• Definition: The integral of a function f (x, y) over a type II region D is

ZZ Z dZ h2 ( y)
f (x, y)dA = f (x, y)d x d y.
c h1 ( y)
D

• Example: Express the following shaded regions both as a Type I region

and a Type II region.
 15

• Example: Consider a lamina occupying the region D bounded by the

parabolas y = 3x 2 and y = 8 + x 2. If the density of the lamina is σ, a
constant, find the centre of mass of the lamina.

Solution:
y

2
y = 8+x

2
y = 3x
x
−2 O 2

The parabolas y = 3x 2 and y = 8 + x 2 intersect when

3x 2 = 8 + x 2

x2 = 4

x = ±2.

By symmetry, x̄ = 0.
16
 17

• Example: Consider a lamina occupying a region D bounded by the line

y = x − 1 and y 2 = 2x + 6. If the pressure is given by p(x, y) = x y 2,
find the total force on the lamina.

Solution: y = x − 1 and y 2 = 2x + 6 intersect at

y2 − 6
y +1=
2
y2 − 2 y − 8 = 0

( y + 2)( y − 4) = 0

y = −2, 4

y
4
2
y =2x+6
y=x−1

O x

−2
18

ZZ Z 4 Z y+1
p(x, y)dA = x y 2d x d y
1 2
−2 2 y −3
D
4
‚ Œ
2  x= y+1
Z 
x 
= y2 dy
−2
2  x= y 2 −3
2
Z 4 2
 4

y y
= y2 + 2 y + 1 − + 3 y2 − 9 d y
−2
2 4
Z4 6

y
= + 2 y4 + y3 − 4 y2 d y
−
−2
8
 7 4
y 2 y5 y4 4 y3
= − + + −
56 5 4 3 −2
 ‹
2048 2048 256 16 64 32
=− + + 64 − − − +4+
7 5 3 7 5 3
3204
=
35
 19

• Example: Consider the double integral

ZZ
I= f (x, y)dA,
D

where D is the region bounded by the curves y = 6x − x 2 and y = 2x +3.

(a) Sketch the region D.
(b) Express I as a iterated integral where the inner integral is with re-
spect to y.
(c) Express I as a iterated integral where the inner integral is with re-
spect to x.
(d) Evaluate the integral I if

1
f (x, y) = p
9− y −2

Solution:
(a)
y
y=2x+3

y=6x−x2

x
O 1 3
20

Z 3Z 6x−x 2
(b) I = f (x, y)d y d x
1 2x+3

(c)

(d)
Z 9Z y−3
2
1
I= p p dxd y
5 3− 9− y 9 − y − 2
Z9   y−3 
1  2
= p x p dy
5 9− y −2 3− 9− y
Z9  ‹
1 y −9 p
= p + 9− y dy
5 9 − y − 2 2
Z9
1 1 p

=− p (9 − y) − 2 9 − y dy
2 5 9− y −2
Z9
1 p
=− 9 − yd y
2 5
3 9

1 (9 − y) 2 
= 3

2 2

5
8
=−
3
 21

5. CHANGING ORDER OF I NTEGRATION

• Example: Consider the integral

Z 1Z 1
sin( y 2)d y d x.
0 x

Sketch the region of integration and compute the integral by reversing

the order of integration.

Solution:
y
y=1

y=x

x
O 1

Z 1Z 1 ZZ
sin( y 2)d y d x = sin( y 2)dA,
0 x
D

where the region D is

D = {(x, y) : 0 ≤ x ≤ 1, x ≤ y ≤ 1}

= {(x, y) : 0 ≤ y ≤ 1, 0 ≤ x ≤ y}
22

Z 1Z 1 Z 1Z y
sin( y 2)d y d x = sin( y 2)d x d y
0 x 0 0
Z 1
= y sin( y 2)d y
0

u = y 2, du = 2 y d y,

y =0 =⇒ u=0

y =1 =⇒ u=1
Z 1Z 1 Z1
1
sin( y 2)d y d x = sin udu
0 x
2 0
• ˜1
1
= − cos u
2 0
1 − cos(1)
=
2
 23

• Example: Consider the integral

Z 1Z 1
p
x 3 + 1d x d y.
p
0 y

Sketch the region of integration and compute the integral by reversing

the order of integration.

Solution:
y
y=1

2
y=x

O x
1

Z 1Z 1 ZZ
p p
x3 + 1d x d y = x 3 + 1dA,
p
0 y
D
where
p
D = {(x, y) : 0 ≤ y ≤ 1, y ≤ x ≤ 1}

= (x, y) : 0 ≤ x ≤ 1, 0 ≤ y ≤ x 2 .
24

Z 1Z 1 p
Z 1Z x2 p
x 3 + 1d x d y = x 3 + 1d y d x
p
0 y 0 0
Z 1p € Š
2
= x3 +1 [ y]0x dx
0
Z1
p
2
= x x 3 + 1d x
0

u = x 3 + 1, du = 3x 2 d x

x =0 =⇒ u=1

x =1 =⇒
u=2
Z 1Z 1 Z2
p 1 1
x 3 + 1d x d y = u 2 du
0
p
y
3 1

2 3 2
• ˜
= u2
9 1
p
4 2−2
=
9
 25

6. POLAR COORDINATES

• Preliminaries: The area of the shaded region is

y

∆θ

O r r +∆ xr

1 1
∆A = (r + ∆r)2∆θ − r 2∆θ ≃ r∆r∆θ .
2 2

• In polar coordinates, if D = {(x, y) : α ≤ θ ≤ β, h1 (θ ) ≤ r ≤ h2(θ )},

then
ZZ Z β Z h2 (θ )
f (x, y)dA = f (r cos θ , r sin θ )r d r dθ
α h1 (θ )
D
26

• Example: Evaluate the integral

p
Z a Z a2 −x 2
(x 2 + y 2 )1/2 d y d x
−a 0

Solution:
y

2 2 2
x +y =a

x
−a O a

p
Z a Z a2 −x 2 Z πZ a
(x 2 + y 2)1/2 d y d x = r 2 d r dθ
−a 0 0 0
Z π  3 a

r
= dθ
0
3 0

πa3
=
3
 27

• Example: Find the area of the region between the circles r = 4 cos θ
and r = 4 sin θ .
Solution:
y

r=2sinθ

D r=2cosθ
2

D
1

O x
28

π
Z 4
Z 4 sin θ
A =2 r d r dθ
0 0
Z π 4 sin θ
4
r 2 
=2 dθ
0
2 0
Z π
4
=16 sin2 θ dθ
0
Z π
4
=8 (1 − cos 2θ )dθ
0
 ‹ π4
sin 2θ 
=8 θ −
2 
0
 ‹
π 1 π
8 − sin
4 2 2
=2π − 4
 29
Rt
• If a
f (x)d x exists for every number t ≥ a, then
Z∞ Z t
f (x)d x := lim f (x)d x
t→∞
a a

provided this limit exists as a finite number.

Rb
• If t
f (x)d x exists for every number t ≤ b, then
Z b Z b
f (x)d x := lim f (x)d x
t→−∞
−∞ t

provided this limit exists as a finite number.

These integrals are called improper integrals. When the corresponding

limit exists, an improper integral is convergent. An improper integral is
divergent when the corresponding limit does not exist.

R∞ Ra
• If both a
f (x)d x and −∞
f (x)d x are convergent, then
Z∞ Za Z∞
f (x)d x := f (x)d x + f (x)d x
−∞ −∞ a
30

• Example: Show that

Z ∞
2 p
e−x d x = π.
−∞

Solution:

REFERENCES
[1] James Stewart, Calculus, Thomson.
[2] C. Henry Edwards and David E. Penney, Calculus, Early Transcendentals, Pearson.