ENGR 233: Applied Advanced Calculus Winter 2023, Lecture Note #4

Instructor: Kiarash Aryankia

Email: kiarash.aryankia@concordia.ca
Reference: D. Zill, Advanced Engineering Mathematics, Jones & Bartlett Learning,
2018.

1 Partial Derivatives
Curve Surface and Level Set
A curve is a one-dimensional object that is defined by a set of mathematical equations. Curves can
be represented graphically and can be used to model a wide range of phenomena in science and
engineering. Some examples of curves include straight lines, parabolas, circles, ellipses, and more
complex shapes like splines (functions defined by polynomials).

In three-dimensional space, a surface is a two-dimensional object that is defined by a set of mathe-

matical equations. It can be represented graphically as well as can be imagined as a thin sheet in 3D
space. Some examples of surfaces include planes, spheres, cylinders, cones, and more complex shapes
like surfaces of revolution (rotating 2d curves), hyperboloids, and tori (like a doughnut). It’s also
possible to have a parametric surface where a set of parametric equations gives the position of a point
on the surface.

Consider z as a function of the pair (x, y), i.e., z = f (x, y). We will call (x, y) independent variables, which
represent the domain of the function, and z is the range of the function.
Now set z a constant value, i.e., z = c; therefore, one has f (x, y) = c, this represents the intersection of
the function and the plane z = c (see Figure 1), which is called level curve. Level surfaces is defined by
w = F (x, y, z) = c

Figure 1: Surface and level curves[taken from the textbook].

page 1 of 7
ENGR 233: Applied Advanced Calculus Winter 2023, Lecture Note #4

Partial Derivatives

Consider a surface F (x, y, z) which is a function of x, y, and z. The partial derivative of F with
respect to its variables is defined by

∂F f (x + ∆x, y, z) − f (x, y, z)
= Fx (x, y, z) = lim
∂x ∆x→0 ∆x
∂F f (x, y + ∆y, z) − f (x, y, z)
= Fy (x, y, z) = lim (1)
∂x ∆y→0 ∆y
∂F f (x, y, z + ∆z) − f (x, y, z)
= Fz (x, y, z) = lim
∂z ∆z→0 ∆z

Using a similar notation, we can define second-order partial derivatives, e.g.,

∂2F
 
∂ ∂F
= Fxx = , (2)
∂x2 ∂x ∂x
or mixed second-order partial derivatives

∂2F
 
∂ ∂F
= Fxy = . (3)
∂x∂y ∂x ∂y

Note: Fxy = Fyx .

Chain Rule in Partial Derivatives

If z = F (u, v) is differentiable function and u = G(x, y) and v = H(x, y) have continuous first partial
derivatives, then
∂z ∂z ∂u ∂z ∂v ∂z ∂z ∂u ∂z ∂v
= + , = + . (4)
∂x ∂u ∂x ∂v ∂x ∂y ∂u ∂y ∂v ∂y

√
Example 1. If z = (ln u2 ) sin vw2 and u = t2 , v = t − 8, w = tan−1 t, find dz/dt.

page 2 of 7
ENGR 233: Applied Advanced Calculus Winter 2023, Lecture Note #4

2 Directional Derivative
Gradient of a Function

It measures the rate of the change of function F1 (x, y, z). It is denoted by ∇F1 and defined as

∂F1 ∂F1 ∂F1

∇F1 (x, y, z) = i+ j+ k (5)
∂x ∂y ∂z

Similarly, for a function F2 (x, y) in 2-space, it is defined as

∂F2 ∂F2
∇F2 (x, y) = i+ j. (6)
∂x ∂y

Directional Derivative

For a differentiable function F1 (x, y, z), the directional derivative (scalar) is defined by

Du F1 (x, y, z) = ∇F1 (x, y, z) · u, (7)

where u is a unit vector. Similarly, for a function F2 (x, y), it is defined as

Du F1 (x, y) = ∇F2 (x, y) · u. (8)

Note: Equations (7) and (8) are inner products of two vectors. From (7) and (8), one can write

Du F1 (x, y, z) = compu ∇F1 (x, y, z), (9)

and
Du F2 (x, y) = compu ∇F2 (x, y). (10)
Note: From (8), if F (x, y) is a differentiable function of x and y and u = cos θi + sin θj, then

Du F (x, y) = ∇F (x, y) · u.

Question: For a given function F , what is the maximum or minimum value of the directional derivative?

Answer:
See (7) and (8), recall the dot product formula; one can write

Du F = ∥∇F ∥∥u∥ cos θ

= ∥∇F ∥ cos θ (11)
≤ ∥∇F ∥.

Thus, the maximum value is ∥∇F ∥.

Du F = ∥∇F ∥∥u∥ cos θ

(12)
≥ −∥∇F ∥.

Thus, the minimum value is −∥∇F ∥.

Question: For a given Function F , when is the maximum or minimum value of the directional derivative
obtained?

page 3 of 7
ENGR 233: Applied Advanced Calculus Winter 2023, Lecture Note #4

Answer:
The maximum value is obtained when cos θ = 1; therefore, u has the same direction as ∇F .
The minimum value is obtained when cos θ = −1; therefore, u has the opposite direction as ∇F .
Note: At a point, if we move along the gradient vector −∇F (or ∇F ) direction, F decreases (or increases)
most rapidly.

If you need to know more, search for “gradient descent”.

2 2 2
Example√ 1. If f (x, y, z) = x + xy + y − x + z + zy, find all points where Du f (x, y) in the direction of
u = (1/ 3)(i + j + k) is zero.
Final answer: 3x+4y+3z= 1

Example 2. Find the directional derivative of the function f (x, y) = y 2 tan2 x;

(a) At π/3, −1


2 , in the direction of the negative y-axis.

(b) At (π/6, 1), in the direction of 3i − 4j.

√
Final answer: (a)3, (b) 8( 3 − 1)/15.

page 4 of 7
ENGR 233: Applied Advanced Calculus Winter 2023, Lecture Note #4

3 Tangent Planes and Normal Lines

Orthogonal Gradient to Level Curve

Let the differentiable function z = F (x, y) be given. For the level curve of the F (x, y) = c which is
parameterized by x = g(t), y = h(t), one has the following equation for its derivative

∂F dx ∂F dy
+ = 0. (13)
∂x dt ∂y dt

Consider the gradient of the function F as (6), and define:

∂F ∂F dx dy
∇F (x, y) = i+ j and r′ (t) = i + j, (14)
∂x ∂y dt dt
Then, one can write
∇F (x, y) · r′ (t) = 0. (15)
Suppose at t = T0 , F (x(T0 ), y(T0 )) = c, from (15), one can conclude that if r′ (t) ̸= 0, then
∇F (x(T0 ), y(T0 )) is orthogonal to tangent vector of r′ (T0 ).

Note that x(T0 ) = g(T0 ), y(T0 ) = h(T0 ).

Orthogonal Gradient to Level Surface

Similarly, let the differentiable surface w = F (x, y, z) is given. For the level curve of the F (x, y, z) = c,
if the differentiable functions x = g(t), y = h(t), z = k(t) are the parametric equations of a curve C on
the surface, one has the following equation for its derivative
∂F dx ∂F dy ∂F dz
+ + = 0, (16)
∂x dt ∂y dt ∂z dt
or    
∂F ∂F ∂F dx dy dz
i+ j+ k · i + j + k = 0. (17)
∂x ∂y ∂z dt dt dt
Using similar definition in (14), one has

∇F (x(T0 ), y(T0 ), z(T0 )) · r′ (t0 ) = 0, (18)

where x(T0 ) = g(T0 ), y(T0 ) = h(T0 ), and z(T0 ) = k(T0 ). If r′ (t) ̸= 0, then
∇F (x(T0 ), y(T0 ), z(T0 )) is orthogonal to tangent vector of r′ (T0 ).

Example 1. Find the level surface of F (x, y, z) = x2 + y 2 + z 2 passing through P (2, 2, 1), and ∇F (see
Figure 2).

page 5 of 7
ENGR 233: Applied Advanced Calculus Winter 2023, Lecture Note #4

Figure 2: The gradient at point P is normal to level surface (Example 1).

Note: The normal vector of a plane is a straight line perpendicular to the tangent.
Recall the tangent line to a graph of a function which is a straight line that touches the function at only
one point.
Suppose P (x0 , y0 , z0 ) be a point on the graph of F (x, y, z) = c, where ∇F ̸= 0. The tangent plane at P is
that plane through P that is normal to ∇F evaluated at P . We have the normal vector and a point on the
tangent plane (the surface), we can write the equation of the plane as ∇F (x0 , y0 , z0 ) · (r − r0 ) = 0.

Tangent Plane

Let P (x0 , y0 , z0 ) be a point on the graph of F (x, y, z) = c, where ∇F ̸= 0 . Then, an equation of the
tangent plane at P is

Fx (x0 , y0 , z0 ) (x − x0 ) + Fy (x0 , y0 , z0 ) (y − y0 ) + Fz (x0 , y0 , z0 ) (z − z0 ) = 0, (19)

where
∂F ∂F ∂F
∇F (x, y, z) = i+ j+ k = Fx i + Fy j + Fz k. (20)
∂x ∂y ∂z

Recall the equation “Planes in 3-space” in Lecture Note #2.

Note: What does happen if ∇F = 0? If ∇F = 0 in (19), you have 0 = 0, therefore, ∇F = 0 is the

trivial solution. A trivial solution is the solution to some equations with a simple structure, e.g., one simple
solution of matrix equation AX = 0 is X = 0, known as a “trivial solution”.

page 6 of 7
ENGR 233: Applied Advanced Calculus Winter 2023, Lecture Note #4

Normal Line to a Surface

Let P (x0 , y0 , z0 ) be a point on the graph of F (x, y, z) = c, where ∇F ̸= 0. Then, the line containing
P (x0 , y0 , z0 ) that is parallel to ∇F (x0 , y0 , z0 ) is called the normal line to the surface at P . The
symmetric equation of the normal line is given by
x − x0 y − y0 z − z0
= = . (21)
Fx Fy Fz

Note: Without loss of generality, one can write the equation of the line as r = ∇F (x, y, z)t + P

Example 2. : Find an equation of the tangent plane and normal line to the graph of z = x2 + y 2 − 4 at
(0, 1, −3).
Final answer: P : 2y − z = 5, L : r = (0, 2t + 1, −t − 3).

Example 3. 7: Find the points on the given surface at which the gradient is parallel to the vector v.

0.5x3 + y + sin(z) = 15; v = 6i + 1j + k (22)

Final answer: (2, 11, 2kπ) and (−2, 19, 2kπ), k ∈ Z.

page 7 of 7