Page 1

MA1511

MA1511 Engineering Calculus

Chapter 3 Vector-valued Functions

3.1 Curves and Motion in Space

Curves of Moving Particles in Space

Consider a particle moving in the three-dimensional space. Its position, 𝑃(𝑥, 𝑦, 𝑧) at time 𝑡 can be
described by three parametric equations

𝑥 = 𝑓(𝑡), 𝑦 = 𝑔(𝑡) and 𝑧 = ℎ(𝑡)

that give the coordinates of 𝑃, where 𝑓, 𝑔 and ℎ are some functions.

The set of all the points (𝑓(𝑡), 𝑔(𝑡), ℎ(𝑡)) is a curve in space that represents the path (or trajectory) of
the particle.
The position vector of P is given by

𝑓(𝑡)
𝐫(𝑡) = (𝑔(𝑡)).
ℎ(𝑡)

We call 𝐫(𝑡) a vector-valued function (or simply, vector function) in one variable. The functions
𝑓(𝑡), 𝑔(𝑡), ℎ(𝑡) are the components of 𝐫(𝑡).

For motion in the two-dimensional space, we simply take the first two components
𝑓(𝑡)
𝐫(𝑡) = ( ).
𝑔(𝑡)

Dr NG Wee Seng
 Page 2
MA1511

An example of a two-dimensional vector function is

𝑎 cos 𝑡
𝐫(𝑡) = ( ) , 𝑎 > 0, for − 𝜋 < 𝑡 ≤ 𝜋,
𝑎 sin 𝑡
or in parametric form,
𝑥 = 𝑎 cos 𝑡, 𝑦 = 𝑎 sin 𝑡
The path represented by the above equations is a circle of radius 𝑎 centered at the origin.

An example of a three-dimensional vector function is

𝑎 + 𝑚𝑡 𝑎 𝑚
𝐫(𝑡) = ( 𝑏 + 𝑛𝑡 ) = (𝑏 ) + 𝑡 ( 𝑛 ), 𝑡 ∈ 𝐑,
𝑐 + 𝑝𝑡 𝑐 𝑝
whose path is a straight line passing through the point (𝑎, 𝑏, 𝑐) and parallel to the vector 𝑚𝐢 + 𝑝𝐣 +
𝑞𝐤 .

Dr NG Wee Seng
 Page 3
MA1511

Derivatives of Vector-valued Functions

For a moving particle whose motion is described by a vector function 𝐫(𝑡), the velocity vector and
acceleration vector are given respectively by

𝑓 ′ (𝑡) 2 𝑓 ′′ (𝑡)
𝑑 𝑑
𝐫 ′ (𝑡) = 𝐫(𝑡) = (𝑔′ (𝑡)) (velocity), 𝐫 ′′ (𝑡) = 2 𝐫(𝑡) = (𝑔′′ (𝑡)) (acceleration)
𝑑𝑡 𝑑𝑡
ℎ′ (𝑡) ℎ′′ (𝑡)
Note that differentiation of a vector function 𝐫(𝑡) is done by componentwise differentiation.

Using a standard calculus argument1, it can be shown that the velocity vector 𝐫 ′ (𝑡) is tangential to the
path at time 𝑡.
The magnitude of the velocity vector,

| 𝐫 ′ (𝑡)| = √(𝑓′(𝑡))2 + (𝑔′(𝑡))2 + (ℎ′(𝑡))2

measures the speed of the particle at time 𝑡.

Remarks
In the two dimensional space,
𝑓(𝑡)
𝐫(𝑡) = ( ), 𝑎 ≤ 𝑡 ≤ 𝑏
𝑔(𝑡)
Derivatives and related quantities can be defined accordingly. For example,
𝑓 ′ (𝑡)
𝐫 ′ (𝑡) = ( ′ )
𝑔 (𝑡)
𝑓′′ (𝑡)
𝐫 ′ ′(𝑡) = ( ′ )
𝑔′ (𝑡)

𝐫(𝑡+ℎ)−𝐫(𝑡)
1
Use the mathematical definition of derivatives, 𝐫 ′ (𝑡) = lim .
ℎ→0 ℎ
Dr NG Wee Seng
 Page 4
MA1511

Remarks (Smooth Curves and Piece-wise Smooth Curves)

𝑓(𝑡)
Geometrically2, a curve 𝐫(𝑡) = (𝑔(𝑡)) , 𝑡 ∈ 𝐼 is said to be smooth it has no sharp corners ( cusps ).
ℎ(𝑡)

Piecewise Smooth Curves

A piecewise smooth curve is made up of a finite number of smooth curves.

In MA1511, we deal only with smooth curves.

2
Mathematically, 𝐶 is smooth if all its component functions are differentiable and 𝐫 ′ (𝑡) ≠ 𝟎 for all 𝑡.
Dr NG Wee Seng
 Page 5
MA1511

Tangent Lines

Let 𝑃(𝑓(𝑡0 ), 𝑔(𝑡0 ), ℎ(𝑡0 )) be a point on the curve represented by

𝑓(𝑡)
𝐫(𝑡) = (𝑔(𝑡)).
ℎ(𝑡)

Since the tangent line to the curve at 𝑃 passes through 𝑃(𝑓(𝑡0 ), 𝑔(𝑡0 ), ℎ(𝑡0 )) and is parallel to the
velocity vector
𝑓 ′ (𝑡0 )
𝐫 ′ (𝑡0 ) = (𝑔′ (𝑡0 )) ,
ℎ′ (𝑡0 )
the equation of this line is given by the following result.

Result 3.1A (Tangent Lines to Parametric Curves)

𝑓(𝑡)
A vector equation of the tangent line to a curve 𝐫(𝑡) = ( 𝑔(𝑡) ) at the point where 𝑡 = 𝑡0 is
ℎ(𝑡)
𝑓(𝑡0 ) 𝑓 ′ (𝑡0 )
𝐫 = (𝑔(𝑡0 )) + 𝑠 (𝑔′ (𝑡0 )) , 𝑠 ∈ 𝐑
ℎ(𝑡0 ) ℎ′ (𝑡0 )

Remarks
1. In the above result, 𝑠 is the parameter. Each value of s corresponds to a specific point on the
tangent line.

Dr NG Wee Seng
 Page 6
MA1511

2. For problems where two or more space curves are involved , different symbols should be used
when finding the intersection of the curves. For example, to find the intersection (if any) of the
following two space curves representing the paths of two moving particles
2𝑡 𝑡
𝐶1 : 𝐫1 (𝑡) = ( ) and 𝐶2 : 𝐫2 (𝑡) = ( )
𝑡 1 − 𝑡

we solve

𝐫1 (𝑠) = 𝐫2 (𝑡)

1
using different symbols (𝑠 and 𝑡) for the time taken. This will lead to the solutions 𝑠 = 3 and
2 2 1
= 3 . It follows that that the two paths cross at the point (3 , 3). Since this occurs when ≠ 𝑡 , we

conclude that the two moving particles do not collide .

Video Example 3.1.1

A particle moves in spaces such that its position at time 𝑡 is.
𝐫(𝑡) = 𝑡𝐢 + 𝑒 1−𝑡 𝐣 − 𝑡 3 k , 𝑡 ≥ 0.
(i) Find the equation of the tangent line to the path at 𝑡 = 1
(ii) Find the speed of the particle at 𝑡 = 1

1 1
Answer: 𝐫(𝑡) = ( 1 ) + 𝑡 (−1), 𝑡 ∈ 𝐑 ; speed = √11
−1 −3

Dr NG Wee Seng
 Page 7
MA1511

Differentiation Rules for Vector-valued Functions (Optional)

Below is a list of differentiation formulae or rules for vector-valued functions that are useful for
differentiating functions involving two or more vector functions, especially when products of vector
functions are involved .
Results (1) to (3) can be proved easily by using the componentwise differentiation of vector functions.
𝑑
Results (4) to (6) are the vector counterparts of the product rule (𝑢(𝑡) ∙ 𝑣(𝑡)) = 𝑢′ (𝑡) ∙ 𝑣(𝑡) +
𝑑𝑡
𝑣′(𝑡) ∙ 𝑢(𝑡) for scalar functions3. Result 7 is the chain rule for vector functions.

Let u(𝑡) and v(𝑡) be two vector-valued functions of 𝑡 , C be a constant vector, 𝑐 be a constant
scalar and 𝑓 be a differentiable scalar function.
𝑑
1. 𝐂=𝟎
𝑑𝑡
𝑑 𝑑
2. 𝑐𝐮(𝑡) = 𝑐 𝐮(𝑡)
𝑑𝑡 𝑑𝑡
𝑑
3. (𝐮(𝑡) + 𝐯(𝑡)) = 𝐮′(𝑡) + 𝐯′(𝑡)
𝑑𝑡
𝑑
4. 𝑓(𝑡)𝐮(𝑡) = 𝑓 ′ (𝑡)𝐮(𝑡) + 𝑓(𝑡)𝐮′(𝑡)
𝑑𝑡
𝑑
5. 𝐮(𝑡) ∙ 𝐯(𝑡) = 𝐮′(𝑡) ∙ 𝐯(𝑡) + 𝐯′(𝑡) ∙ 𝐮(𝑡)
𝑑𝑡
𝑑
6. 𝐮(𝑡) × 𝐯(𝑡) = 𝐮′(𝑡) × 𝐯(𝑡) + 𝐮(𝑡) × 𝐯′(𝑡)
𝑑𝑡
𝑑
7. 𝐮(𝑓(𝑡)) = 𝑓 ′ (𝑡) 𝐮′(𝑓(𝑡))
𝑑𝑡

3
See the proofs of (4) and (5) in LumiNUS > Files > Supplementary Slides folder
Dr NG Wee Seng
 Page 8
MA1511

Two Commonly Encountered Parametric Curves

1. Line Segments

The line segment joining two distinct points, 𝐴(𝑥1 , 𝑦1 , 𝑧1 ) and 𝐵(𝑥2 , 𝑦2 , 𝑧2 ) has parametric
representations
𝑥1 𝑥2 (1 − 𝑡)𝑥1 + 𝑡𝑥2
𝑦 𝑦
𝐫(𝑡) = (1 − 𝑡) ( 1 ) + 𝑡 ( 2 ) = ((1 − 𝑡)𝑦1 + 𝑡𝑦2 ) , 0 ≤ 𝑡 ≤ 1.
𝑧1 𝑧2 (1 − 𝑡)𝑧1 + 𝑡𝑧2

To see this, we first write

⃗⃗⃗⃗⃗ = 𝑡𝐴𝐵
𝐴𝑃 ⃗⃗⃗⃗⃗ for some 0 ≤ 𝑡 ≤ 1

This gives

⃗⃗⃗⃗⃗ − ⃗⃗⃗⃗⃗⃗
𝑂𝑃 ⃗⃗⃗⃗⃗ − ⃗⃗⃗⃗⃗⃗
𝑂𝐴 = 𝑡( 𝑂𝐵 𝑂𝐴 )
and hence,

𝑂𝑃 = (1 − 𝑡)⃗⃗⃗⃗⃗⃗
𝐫(𝑡) = ⃗⃗⃗⃗⃗ 𝑂𝐴 + 𝑡 ⃗⃗⃗⃗⃗⃗⃗
𝑂𝐵
as required.
Note that 𝑡 = 0 and 𝑡 = 1 give the points 𝐴 and 𝐵 respectively.

Remark
In a two-dimensional space, the line segment joining two distinct points 𝐴(𝑥1 , 𝑦1 ) and
𝐵(𝑥2 , 𝑦2 ) has parametric representations
(1 − 𝑡)𝑥1 + 𝑡𝑥2
𝐫(𝑡) = ( ),0 ≤ 𝑡 ≤ 1
(1 − 𝑡)𝑦1 + 𝑡𝑦2

Dr NG Wee Seng
 Page 9
MA1511

2. Circles (In a 2-Dimensional Space)

A circle of radius 𝑟 centered at the origin (0, 0) has parametric representations

𝑟 cos 𝑡
𝐫(𝑡) = ( ) , 0 ≤ 𝑡 ≤ 2𝜋
𝑟 sin 𝑡
where 𝑡 = 0 and t = 2𝜋 both correspond to the point (1, 0) and the direction of increasing
𝑡 is an anti-clockwise traversal of the circle from the initial point (1, 0) back to the same point,
(1, 0)

Remark
The range of values of 𝑡 has to be adjusted when C is a sector of a circle.
For example, if C is a semi-circle from (0, 1) to (0, -1) in an anti-clockwise sense,
𝜋 3𝜋
then ≤𝑡≤ (see the above diagram)
2 2

Dr NG Wee Seng
 Page 10
MA1511

3.2 Integrals of Vector – valued Functions

Let
𝑓(𝑡)
𝐫(𝑡) = (𝑔(𝑡)) = (𝑓(𝑡))𝐢 + (𝑔(𝑡))𝐣 + (ℎ(𝑡))𝐤
ℎ(𝑡)
Indefinite Integrals

We define the indefinite integral ∫ 𝐫(𝑡)𝑑𝑡 in terms of its component functions 𝑓, 𝑔 and ℎ by

∫ 𝐫(𝑡)𝑑𝑡 = (∫ 𝑓(𝑡)𝑑𝑡)𝐢 + (∫ 𝑔(𝑡)𝑑𝑡)𝐣 + (∫ ℎ(𝑡)𝑑𝑡)𝐤 .

For example, if
𝐫(𝑡) = 2𝑡 𝐢 + (sin 𝑡)𝐣 + (cos 𝑡)𝐤
then,

∫ 𝐫(𝑡)𝑑𝑡 = ( ∫ 2𝑡𝑑𝑡 )𝐢 + (∫ sin 𝑡 𝑑𝑡)𝐣 + (∫ cos 𝑡 𝑑𝑡)𝐤

=( 𝑡 2 + 𝐶1 )𝐢 + (− cos 𝑡 + 𝐶2 ) 𝐣 + (sin 𝑡 + 𝐶3 )𝐤
where 𝐶1 , 𝐶2 and 𝐶3 are arbitrary constants. Note that we do not use the same integration constant
for the three integrals.

Given the relation between the position vector 𝐫(𝑡) and the velocity vector, 𝐫 ′ (𝑡), we can integrate
the velocity vector to get the position vector. Similarly, we integrate the acceleration vector to obtain
the velocity vector.

Definite Integrals
𝑏
We define the definite integral ∫𝑎 𝐫(𝑡)𝑑𝑡 in terms of its component functions 𝑓, 𝑔 and ℎ by

𝑏 𝑏 𝑏 𝑏
∫𝑎 𝐫(𝑡)𝑑𝑡 = (∫𝑎 𝑓(𝑡)𝑑𝑡) 𝐢 + (∫𝑎 𝑔(𝑡)𝑑𝑡) 𝐣 + (∫𝑎 ℎ(𝑡)𝑑𝑡) 𝐤 .

Dr NG Wee Seng
 Page 11
MA1511

Video Example 3.2.1

A particle moves in space in such a way that its acceleration vector is
𝑡
a(𝑡) = ( 𝑒 𝑡 )
sin 𝑡
The initial (𝑡 = 0) position and initial velocity vector are (1, 0, 1) and 2i .Find the position
vector r(𝑡) of the particle.

𝑡3
+ 2𝑡 + 1
6
Answer: 𝐫(𝑡) = ( )
𝑒𝑡 − 𝑡 − 1
− sin 𝑡 + 𝑡 + 1

Dr NG Wee Seng
 Page 12
MA1511

3.3 Arc Length

The length , 𝐿 of a smooth curve defined by the vector function
𝑓(𝑡)
𝐫(𝑡) = (𝑔(𝑡)), 𝑎 ≤ 𝑡 ≤ 𝑏,
ℎ(𝑡)

and traced exactly once as 𝑡 increases from 𝑡 = 𝑎 to 𝑡 = 𝑏, is given by

Result 3.3A (Length of Smooth Curve)
𝑏

𝐿 = ∫ √(𝑓′(𝑡))2 + (𝑔′(𝑡))2 + (ℎ′(𝑡))2 𝑑𝑡

𝑎

The mathematical proof of Result 3.3A is beyond the scope of this course4. Nonetheless, we can
interpret the arc length as the distance travelled by the particle from 𝑡 = 𝑎 to 𝑡 = 𝑏 by observing
that the integrand in the above formula, namely

√(𝑓′(𝑡))2 + (𝑔′(𝑡))2 + (ℎ′(𝑡))2

is the speed of the particle at time t (See Chapter 3.1). Hence, we are integrating the speed function
over time, which gives the distance travelled.

Remarks
1. For a curve in a two-dimensional space:
𝑓(𝑡)
𝐫(𝑡) = ( ), 𝑎 ≤ 𝑡 ≤ 𝑏,
𝑔(𝑡)
𝑏

𝐿 = ∫ √(𝑓′(𝑡))2 + (𝑔′(𝑡))2 𝑑𝑡
𝑎

2 For a curve defined explicitly by 𝑦 = 𝑓(𝑥) , if we represent C by the following pair of

parametric equations
𝑥 = 𝑡, 𝑦 = 𝑓(𝑡)
where 𝑎 ≤ 𝑡 ≤ 𝑏 for some real numbers a and 𝑏 , a < b, then the above formula gives the
length of C

4
Integral formulae such as Result 3.3A are proved using the concept of limits of Riemann sums
http://mathworld.wolfram.com/RiemannSum.html

Dr NG Wee Seng
 Page 13
MA1511

𝐿 = ∫ √1 + (𝑓′(𝑡))2 𝑑𝑡
𝑎

To illustrate the arc length formula, we consider a two-dimensional unit circle centered at the origin
(0, 0) which can be parametrised by the vector function
cos 𝑡
𝐫(𝑡) = ( ) , 0 ≤ 𝑡 ≤ 2𝜋.
sin 𝑡
The length of the circumference of this circle, which we know is 2π, can be found by the above
formula :
2𝜋 2𝜋

L = ∫ √(− sin 𝑡) 2 + (cos 𝑡)2 𝑑𝑡 = ∫ √1𝑑𝑡 = 2𝜋.

0 0

Video Example 3.3.1

A particle moves along the helix
r(𝑡) = (𝑐𝑜𝑠 𝑡)i + (sin t) j + t k
from 𝑡 = 0 to 𝑡 = 2𝜋.
How long is the path?
Answer: 2√2𝜋

Dr NG Wee Seng
 Page 14
MA1511

3.4 Line Integrals

Geometrical Interpretation of Line Integrals (Optional)
As a motivation for the line integrals that we about to define, we recall that the area between the curve
𝑦 = 𝑓(𝑥) of a non-negative function 𝑓 and the 𝑥 −axis for 𝑎 ≤ 𝑥 ≤ 𝑏 is given by the definite
integral
𝑏

∫ 𝑓(𝑥)𝑑𝑥.
𝑎

We can think of this integral is a summation process in which areas of rectangles under the curve are
summed via a limiting process5. A typical rectangle (the shaded rectangle below) is one whose height
is 𝑓(𝑡), 𝑎 ≤ 𝑡 ≤ 𝑏 and whose width is the interval length, ∆𝑡 . The area of this rectangle is therefore
𝑓(𝑡)∆𝑡 , giving rise to the above integral.

We would like to extend the above concept to the integration of a function 𝑓(𝑥, 𝑦) over a smooth
curve 𝐶 represented by a vector function 𝐫(𝑡) = 𝑥(𝑡) 𝐢 + 𝑦(𝑡) 𝐣 , 𝑎 ≤ 𝑡 ≤ 𝑏 . When restricted to
the curve 𝐶 as its domain, the function 𝑓(𝑥, 𝑦) will take values 𝑓(𝑥(𝑡), 𝑦(𝑡)) for 𝑎 ≤ 𝑡 ≤ 𝑏, as
illustrated in the diagram below.

5
using the concept of limits of Riemann sums http://mathworld.wolfram.com/RiemannSum.html.
Dr NG Wee Seng
 Page 15
MA1511

Like the integral of 𝑓(𝑥) over 𝑎 ≤ 𝑥 ≤ 𝑏, the integral of 𝑓(𝑥, 𝑦) over 𝐶 : 𝐫(𝑡) = 𝑥(𝑡) 𝐢 + 𝑦(𝑡) 𝐣 ,
𝑎 ≤ 𝑡 ≤ 𝑏 involves summing areas of “rectangles” formed by taking the product of 𝑓(𝑥(𝑡), 𝑦(𝑡)) and
𝑑𝑠, the length of the small arc of the curve defined on [𝑡, 𝑡 + ∆𝑡] and taking limit ∆𝑡 → 0. 6.

Line Integrals

It turns out that the integral of a function 𝑓(𝑥, 𝑦) over a smooth curve 𝐶 is given by

𝑏
2 2
∫ 𝑓(𝑥(𝑡), 𝑦(𝑡))√(𝑥′ (𝑡)) + (𝑦′ (𝑡)) 𝑑𝑡
𝑎

2 2
where √(𝑥 ′ (𝑡)) + (𝑦 ′ (𝑡)) 𝑑𝑡 corresponds to the term "𝑑𝑠" in the above illustration.

We call this the line integral of 𝒇 along 𝑪. A standard notation for this integral is

∫ 𝑓(𝑥, 𝑦) 𝑑𝑠
𝐶

We summarise the above in Result 3.4A.

Result 3.4A (Line Integrals over Two-dimensional Smooth Curves)

If 𝑓(𝑥, 𝑦) is defined on a smooth curve 𝐶 : 𝐫(𝑡) = 𝑥(𝑡) 𝐢 + 𝑦(𝑡) 𝐣 , 𝑎 ≤ 𝑡 ≤ 𝑏 , the line

integral of 𝑓 along 𝐶 , denoted by ∫𝐶 𝑓 (𝑥, 𝑦) 𝑑𝑠 , is given by

𝑏
2 2
∫ 𝑓(𝑥, 𝑦) 𝑑𝑠 = ∫ 𝑓(𝑥(𝑡), 𝑦(𝑡))√(𝑥 ′ (𝑡)) + (𝑦 ′ (𝑡)) 𝑑𝑡
𝐶
𝑎

6
using the concept of limits of Riemann sums http://mathworld.wolfram.com/RiemannSum.html
Dr NG Wee Seng
 Page 16
MA1511

We can extend naturally the definition of line integrals to functions in three variables, 𝑓(𝑥, 𝑦, 𝑧)
over a three-dimensional smooth curve : 𝐫(𝑡) = 𝑥(𝑡) 𝐢 + 𝑦(𝑡) 𝐣 +𝑧(𝑡) 𝐤 , 𝑎 ≤ 𝑡 ≤ 𝑏 :

Result 3.4B (Line Integrals over Three-dimensional Smooth Curves)

If 𝑓(𝑥, 𝑦, 𝑧) is defined on a smooth curve 𝐶 : 𝐫(𝑡) = 𝑥(𝑡) 𝐢 + 𝑦(𝑡) 𝐣 + 𝑧(𝑡) 𝐤 , 𝑎 ≤ 𝑡 ≤

𝑏 , the line integral of 𝑓 along 𝐶 , denoted by ∫𝐶 𝑓 (𝑥, 𝑦, 𝑧) 𝑑𝑠 , is given by

∫ 𝑓(𝑥(𝑡), 𝑦(𝑡), 𝑧(𝑡))√(𝑥 ′ (𝑡))2 + (𝑦 ′ (𝑡))2 + (𝑧′(𝑡))2 𝑑𝑡

𝑎

Remark

When 𝑓(𝑥, 𝑦) = 1, the two-dimensional line integral ∫𝐶 𝑓(𝑥, 𝑦) 𝑑𝑠 gives the length of the curve C.
(See chapter 3.3)

Similarly, when 𝑓(𝑥, 𝑦, 𝑧) = 1, the three-dimensional line integral ∫𝐶 𝑓(𝑥, 𝑦, 𝑧) 𝑑𝑠 gives the length
of the curve 𝐶. (See chapter 3.3)

Dr NG Wee Seng
 Page 17
MA1511

Line Integrals over Two-dimensional Piecewise Curves

Let 𝐶 be a union of smooth curves 𝐶1 , 𝐶2 , … , 𝐶𝑛 joined together in a continuous manner (that is, the
terminal point of 𝐶1 = the initial point of 𝐶2 , and so on ).

We define the line integral of 𝑓 over 𝐶 as the sum of the individual line integrals

∫ 𝑓 𝑑𝑠 = ∫ 𝑓 𝑑𝑠 + ∫ 𝑓 𝑑𝑠 + ⋯ + ∫ 𝑓 𝑑𝑠.
𝐶 𝐶1 𝐶2 𝐶𝑛

Orientation of Curves
Given a curve C represented by 𝐫(𝑡) = 𝑥(𝑡) 𝐢 + 𝑦(𝑡)𝐣 , 𝑎 ≤ 𝑡 ≤ 𝑏, the direction of the movement
along the curve corresponding to increasing (decreasing) values of 𝒕 gives what we call a positive
(negative) orientation of the curve.

For example, if C is the line joining 𝐴(0, 0) to 𝐵(1, 1), the representation

r(𝑡) = 𝑡 𝐢 + 𝑡 𝐣 , 0 ≤ 𝑡 ≤ 1

corresponds to a positive orientation, while the representation

r(𝑡) = (1 − 𝑡) 𝐢 + (1 − 𝑡) 𝐣 , 0 ≤ 𝑡 ≤ 1

corresponds to a negative orientation (since 𝑡 = 1 gives the point 𝐴 and 𝑡 = 0 gives the point 𝐵)

In computing line integrals, we assume 𝐶 is positively oriented.

Dr NG Wee Seng
 Page 18
MA1511

Let us focus on the use of the line integral formula. We will consider an application of the line integrals
in Tutorial 3.

Video Example 3.4.1

Find the line integral   yx  1 ds

C
where 𝐶 is the upper half of the unit circle 𝑥 2 + 𝑦 2 = 1
from (1, 0) to (-1, 0)
Answer: 𝜋

Video Example 3.4.2

Find the line integral C

x  3 y 2  z ds where 𝐶 is line segment joining (0, 0, 0) to (1, 1, 1)
Answer: 2√3

Dr NG Wee Seng
 Page 19
MA1511

3.5 Parametric Surfaces

Parametric Surfaces
So far, our definition of a surface in three-dimensions has been the graph of a function 𝑓 defined
explicitly by

𝑧 = 𝑓(𝑥, 𝑦).
We now introduce a more general way of representing a surface using parametric equations.

Consider a vector function 𝐫(𝑢, 𝑣) of two variables 𝑢 and 𝑣, defined on some domain 𝐷 in a two-
dimensional space by

𝐫(𝑢, 𝑣) = 𝑥(𝑢, 𝑣)𝐢 + 𝑦(𝑢, 𝑣)𝐣 + 𝑧(𝑢, 𝑣)𝐤 + 𝑧(𝑢, 𝑣)𝐤

For each (𝑢, 𝑣) in 𝐷, 𝐫(𝑢, 𝑣) represents the position vector of a point in space. These points constitute
a surface.
We call this a parametric surface.

Parametrization of Surfaces Defined Explicitly by 𝒛 = 𝒈(𝒙, 𝒚)

It is easy to see that surfaces defined explicitly by 𝑧 = 𝑔(𝑥, 𝑦) can be parametrized by the equations

𝑥 = 𝑢, 𝑦 = 𝑣 and 𝑧 = 𝑔(𝑢, 𝑣)

Dr NG Wee Seng
 Page 20
MA1511

A sphere of radius 𝑎 and centered at the origin has Cartesian equation

𝑥 2 + 𝑦 2 + 𝑧 2 = 𝑎2 .
The following video example gives its parametric representation.
Video Example 3.5.1

Show that the following parametric equations represent a sphere in the three-dimensional space.
𝑥 = 𝑎𝑠𝑖𝑛 𝛼 𝑐𝑜𝑠 𝛽, 𝑦 = 𝑎𝑠𝑖𝑛 𝛼 𝑠𝑖𝑛 𝛽, 𝑧 = 𝑎𝑐𝑜𝑠 𝛼
0 ≤ 𝛼 ≤ 𝜋, 0 ≤ 𝛽 ≤ 2𝜋.

Dr NG Wee Seng
 Page 21
MA1511

Normal Vector to Parametric Surfaces

We recall Result 1.4A that a vector equation of the tangent plane to a surface defined explicitly by
𝑧 = 𝑔(𝑥, 𝑦) at the point (𝑥, 𝑦) = (𝑎, 𝑏) is

𝑔𝑥 (𝑎, 𝑏) 𝑔𝑥 (𝑎, 𝑏) 𝑎
𝐫  (𝑔𝑦 (𝑎, 𝑏)) = (𝑔𝑦 (𝑎, 𝑏))  ( 𝑏 )
−1 −1 𝑔(𝑎, 𝑏)
𝑥
where 𝐫 = (𝑦) is the position vector of any piont on the plane.
𝑧

For a surface defined parametrically by

𝐫(𝑢, 𝑣) = 𝑥(𝑢, 𝑣)𝐢 + 𝑦(𝑢, 𝑣)𝐣 + 𝑧(𝑢, 𝑣)𝐤 , (𝑢, 𝑣) ∈ 𝐷,

we define the vectors of partial derivatives, 𝐫𝒖 , 𝐫𝒗 defined by

𝐫𝑢 = 𝑥𝑢 𝐢 + 𝑦𝑢 𝐣 + 𝑧𝑢 𝐤

𝐫𝑣 = 𝑥𝑣 𝐢 + 𝑦𝑣 𝐣 + 𝑧𝑣 𝐤

The parametric surface is said to be smooth if 𝐫𝒖 × 𝐫𝒗 ≠ 𝟎 for all (𝑢, 𝑣) in 𝐷. In this course, we deal
only with smooth surfaces.

Given a smooth surface 𝐫(𝑢, 𝑣) and a point P where (𝑢, 𝑣) = (𝑢0 , 𝑣0 ) , the vectors 𝐫𝑢 (𝑢0 , 𝑣0 ) and
𝐫𝑣 (𝑢0 , 𝑣0 ) are parallel to the tangent plane to the surface at P. It follows that the vector

(𝐫𝒖 × 𝐫𝒗 )(𝑢0 , 𝑣0 ) is a normal vector to the tangent plane at P.

Hence, a vector equation of the tangent plane at P can be determined.

Dr NG Wee Seng
 Page 22
MA1511

Result 3.5A
Let 𝑃 be the point on the parametric surface

r(𝑢, 𝑣) = 𝑥(𝑢, 𝑣)𝐢 + 𝑦(𝑢, 𝑣)𝐣 + 𝑧(𝑢, 𝑣)𝐤 , (𝑢, 𝑣) ∈ 𝐷

with position vector r(𝑢0 , 𝑣0 ) = 𝑥0 𝐢 + 𝑦0 𝐣 + 𝑧0 𝐤 and let n = (𝐫𝒖 × 𝐫𝒗 ) (𝑢0 , 𝑣0 ) be the vector
𝐫𝒖 × 𝐫𝒗 evaluated at (𝑢0 , 𝑣0 ).

A vector equation of the tangent plane at 𝑃 to the surface is given by

𝑥0
r  n= (𝑦0 )  𝐧
𝑧0
𝑥
where 𝐫 = (𝑦) is the position vector of any point on the tangent plane.
𝑧

Video Example 3.5.2

Find a Cartesian equation of the tangent plane to the surface with parametric equations

𝑥 = 𝑢2 , 𝑦 = 𝑣 2 , 𝑧 = 𝑢 + 2𝑣

at the point where 𝑢 = 1 and 𝑣 = 1.

Answer: 𝑥 + 2𝑦 − 2𝑧 + 3 = 0

End of Chapter 3

“I can calculate the motion of heavenly bodies, but not the madness of people.”

– Isaac Newton

Dr NG Wee Seng