13.

1 1

Chapter 13
PARTIAL DERIVATIVES
(Follows from [1] Anton, …)

13.1 FUNCTIONS OF TWO OR MORE VARIABLES

NOTATION AND TERMINOLOGY
𝑧 = 𝑓(𝑥, 𝑦)
𝑢 = 𝑓(𝑥, 𝑦, 𝑧)
𝑢 = 𝑓(𝑥1 , 𝑥2 , ⋯ , 𝑥𝑛 )
13.1.1 Definition A function f of two variables, x and y, is a rule that assigns a
unique real number 𝑓(𝑥, 𝑦) to each point (𝑥, 𝑦) in some set D in the xy-plane.

13.1.2 Definition A function f of three variables, x, y, and z, is a rule that assigns

a unique real number 𝑓(𝑥, 𝑦, 𝑧) to each point (𝑥, 𝑦, 𝑧) in some set D in three
dimensional space.

Example 1 Sketch the natural domain of

𝑓(𝑥, 𝑦) = √𝑦 + 1 + ln(𝑥 2 − 𝑦).

𝑦 + 1 ≥ 0 & 𝑥2 − 𝑦 > 0

=>
𝑦 ≥ −1 & 𝑦 < 𝑥 2

Example 2 Find the natural domain of 𝑓 (𝑥, 𝑦, 𝑧) = √1 − 𝑥 2 − 𝑦 2 − 𝑧 2 .

 1 − 𝑥 2 − 𝑦 2 − 𝑧 2 ≥ 0 => 𝑥 2 + 𝑦 2 + 𝑧 2 ≤ 1

Therefore, the natural domain of 𝑓 consists of all points on or within the

sphere
𝑥 2 + 𝑦 2 + 𝑧 2 = 1.

Graphs of Functions of Two Variables

Recall that for a function f of one variable, the graph of f(x) in the xy-plane was
defined to be the graph of the equation y = f(x). Similarly, if f is a function of two
variables, we define the graph of f(x, y) in xyz-space to be the graph of the equation
z = f(x, y). In general, such a graph will be a surface in 3-space.

Example 3 In each part, describe the graph of the function in an xyz-

coordinate system.
1
(a) 𝑓(𝑥, 𝑦) = 1 − 𝑥 − 𝑦
2
(b) 𝑓(𝑥, 𝑦) = √1 − − 𝑦 2 𝑥2
(c) 𝑓(𝑥, 𝑦) = −√𝑥 2 + 𝑦 2

Solution (a) By definition, the graph of the given function is the graph of the
1
equation 𝑧 = 1 − 𝑥 – 𝑦 which is a plane. A triangular portion of the plane can
2
be sketched by plotting the intersections with the coordinate axes and joining
them with line segments (Figure 13.1.2a).

Figure 13.1.2
Solution (b) After squaring both sides, this can be rewritten as
𝑥2 + 𝑦2 + 𝑧2 = 1
which represents a sphere of radius 1, centered at the origin. Since (b) imposes the
added condition that z ≥ 0, the graph is just the upper hemisphere (Figure 13.1.2b).

Solution (c)
After squaring, we obtain 𝑧 2 = 𝑥 2 + 𝑦 2 which is the equation of a circular cone
(see Table 11.7.1). Since (c) imposes the condition that z ≤ 0, the graph is
just the lower nappe of the cone (Figure 13.1.2c).

LEVEL CURVES
We are all familiar with the topographic (or contour) maps in which a three-
dimensional landscape, such as a mountain range, is represented by two-
dimensional contour lines or curves of constant elevation. Consider, for example,
the model hill and its contour map shown in Figure 13.1.3. The contour map is
constructed by passing planes of constant elevation through the hill, projecting the
resulting contours onto a flat surface, and labeling the contours with their
elevations. In Figure 13.1.3, note how the two gullies appear as indentations in the
contour lines and how the curves are close together on the contour map where the
hill has a steep slope and become more widely spaced where the slope is gradual.

Figure 13.1.3
 Figure 13.1.4
Contour maps are also useful for studying functions of two variables. If the surface
z = f(x, y) is cut by the horizontal plane z = k, then at all points on the intersection
we have f(x, y) = k. The projection of this intersection onto the xy-plane is called
the level curve of height k or the level curve with constant k (Figure 13.1.4). A set
of level curves for z = f(x, y) is called a contour plot or contour map of f.

Example 4 The graph of the function 𝑓(𝑥, 𝑦) = 𝑦 2 − 𝑥 2 in xyz-space is the

hyperbolic paraboloid (saddle surface) shown in Figure 13.1.5a. The level curves
have equations of the form 𝑦 2 − 𝑥 2 = 𝑘. For k > 0 these curves are hyperbolas
opening along lines parallel to the y-axis; for k < 0 they are hyperbolas opening
along lines parallel to the x-axis; and for k = 0 the level curve consists of the
intersecting lines y + x = 0 and y − x = 0 (Figure 13.1.5b).

Figure 13.1.5

Example 5 Sketch the contour plot of 𝑓(𝑥, 𝑦) = 4𝑥 2 + 𝑦 2 using level curves

of height k = 0, 1, 2, 3, 4, 5.
Solution. The graph of the surface 𝑧 = 4𝑥 2 + 𝑦 2 is the paraboloid shown in the
left part of Figure 13.1.6, so we can reasonably expect the contour plot to be a
family of ellipses centered at the origin. The level curve of height k has the equation
4𝑥 2 + 𝑦 2 = 𝑘. If k = 0, then the graph is the single point (0, 0). For k > 0 we can
rewrite the equation as
𝑥2 𝑦2
+ =1
𝑘/4 𝑘
which represents a family of ellipses with x-intercepts ±√𝑘/2 and y-intercepts
±√𝑘. The contour plot for the specified values of k is shown in the right part of
Figure 13.1.6.

Exercise 13.1
23–26 Sketch the domain of f. Use solid lines for portions of the boundary included
in the domain and dashed lines for portions not included.

35–42 Sketch the graph of f.

43–44 In each part, select the term that best describes the level
curves of the function f. Choose from the terms lines, circles,
noncircular ellipses, parabolas, or hyperbolas.

13.2 Limits and Continuity

Limits along Curves

There are infinitely many different curves along which one point can approach
another (Figure 13.2.1). Our first objective in this section is to define the limit of
𝑓(𝑥, 𝑦) as (x, y) approaches a point (𝑥0 , 𝑦0 ) along a curve C (and similarly for
functions of three variables).

Figure 13.2.1
In these formulas the limit of the function of t must be treated as a one-sided limit
if (𝑥0 , 𝑦0 ) or (𝑥0 , 𝑦0 , 𝑧0 ) is an endpoint of C.

Example 1 Given that

𝑥𝑦
𝑓(𝑥, 𝑦) = − .
𝑥 2 + 𝑦2
Find the limit along
(a) the x-axis (b) the y-axis (c) the line 𝑦 = 𝑥
(d) the line 𝑦 = −𝑥 (e) the parabola 𝑦 = 𝑥 2
Solution (a). The x-axis has parametric equations, 𝑥 = 𝑡, 𝑦 = 0, with (0, 0)
corresponding to t = 0, so

Solution (b). The y-axis has parametric equations x = 0, y = t , with

(0, 0) corresponding to t = 0, so

Solution (c) The line y = x has parametric equations x = t, y = t , with (0, 0)

corresponding to t = 0, so

Solution (d). The line 𝑦 = −𝑥 has parametric equations x = t, y = −t , with

(0, 0)corresponding to t = 0, so

Solution (e). The parabola 𝑦 = 𝑥 2 has parametric equations x = t, y = t , with

(0, 0) corresponding to t = 0, so

GENERAL LIMITS OF FUNCTIONS OF TWO VARIABLES

The statement
is intended to convey the idea that the value of 𝑓(𝑥, 𝑦) can be made as close as
we like to the number L by restricting the point (x, y) to be sufficiently close to
(but different from) the point (𝑥0 , 𝑦0 ).
Example 2

Relationships between General Limits and Limits along smooth Curves

Example 3
CONTINUITY

Example 4
Example 5

Solution

Example 6
Example 7

Solution

Exercise 13.2
13.3 PARTIAL DERIVATIVES

Partial Derivatives of Functions of Two Variables

Partial Derivatives Viewed As Rates of Change and Slopes

13.3.1 Definition If z = f(x, y) and (𝑥0 , 𝑦0 ) is a point in the domain of f, then the
partial derivative of f with respect to x at (𝑥0 , 𝑦0 ) [also called the partial derivative
of z with respect to x at (𝑥0 , 𝑦0 )] is the derivative at 𝑥0 of the function that results
when 𝑦 = 𝑦0 is held fixed and x is allowed to vary. This partial derivative is
denoted by 𝑓𝑥 (𝑥0 , 𝑦0 ) and is given by

(1)

Similarly, the partial derivative of f with respect to y at (𝑥0 , 𝑦0 ) [also called the
partial derivative of z with respect to y at (𝑥0 , 𝑦0 )] is the derivative at 𝑦0 of the
function that results when 𝑥 = 𝑥0 is held fixed and y is allowed to vary. This
partial derivative is denoted by 𝑓𝑦 (𝑥0 , 𝑦0 ) and is given by

(2)

Example 1

For

We obtain

Also since,

We obtain,

The Partial Derivative Functions

Formulas (1) and (2) define the partial derivatives of a function at a specific point
(𝑥0 , 𝑦0 ). However, often it will be desirable to omit the subscripts and think of the
partial derivatives as functions of the variables x and y. These functions are
Example 2

and hence compute

Solution

Partial Derivative Notation

Example 3

Solution

Example 5

Solution.
(a)

(b)
Home work
Example 7, Example 8.
Partial Derivatives and Continuity
In contrast to the case of functions of a single variable, the existence of
partial derivatives for a multivariable function does not guarantee the
continuity of the function. This fact is
shown in the following example.

Example 9 Let

(3)

Except at (0, 0), the partial derivatives of f are

(4 – 5)
It is not evident whether f has partial derivatives at (0, 0), and if so, what the values
of those derivatives are. To answer that question we will have to use the definitions
of the partial derivatives (Definition 13.3.1). Applying Formulas (1) and (2) to (3)
we obtain
This shows that f has partial derivatives at (0, 0) and the values of both partial
derivatives are 0 at that point.
Solution (b) We saw in Example 3 of Section 13.2 that

does not exist. Thus, f is not continuous at (0, 0).

Partial Derivatives of Functions With More Than Two Variables

For a function f(x, y, z) of three variables, there are three partial derivatives:
𝑓𝑥 (𝑥, 𝑦, 𝑧), 𝑓𝑦 (𝑥, 𝑦, 𝑧), 𝑓𝑧 (𝑥, 𝑦, z)
The partial derivative 𝑓𝑥 is calculated by holding y and z constant and
differentiating with respect to x. For 𝑓𝑦 the variables x and z are held constant, and
for 𝑓𝑧 the variables x and y are held constant. If a dependent variable
𝑤 = 𝑓(𝑥, 𝑦, 𝑧)
is used, then the three partial derivatives of f can be denoted by
𝜕𝑤 𝜕𝑤 𝜕𝑤
, , .
𝜕𝑥 𝜕𝑦 𝜕𝑧
Home work
Example 10, Example 11.
Example 12 Find the second-order partial derivatives of
Solution. We have
EXERCISE SET 13.3
3–10 Evaluate the indicated partial derivatives
15.5 THE CHAIN RULE

Example 1 Suppose that

𝑧 = 𝑥 2 𝑦, 𝑥 = 𝑡 2 , 𝑦 = 𝑡 3
Use the chain rule to find dz/dt, and check the result by expressing z as a function
of t and differentiating directly.
Solution. By the chain rule [Formula (5)],

Example 2 H.W.
Example 3 Given that
.
Find

Solution.
 .
Example 4

Solution.
 (14)
Example 7
Given that
𝑥 3 + 𝑦 2 𝑥 − 3 = 0.
Find dy/dx using (14), and check the result using implicit differentiation.
Solution.
Example 8
Consider the sphere 𝑥 2 + 𝑦 2 + 𝑧 2 = 1. Find ∂z/∂x and ∂z/∂y at the
point (2/3, 1/3, 2/3)

Solution. By Theorem 13.5.4 with 𝑓(𝑥, 𝑦, 𝑧) = 𝑥 2 + 𝑦 2 + 𝑧 2 ,

EXERCISE SET 13.5
1–5 Use an appropriate form of the chain rule to find dz/dt.

7–10 Use an appropriate form of the chain rule to find 𝑑𝑤/𝑑𝑡.

17–22 Use appropriate forms of the chain rule to find 𝜕𝑧/𝜕𝑢 and 𝜕𝑧/𝜕𝑣.
13.6 DIRECTIONAL DERIVATIVES AND GRADIENTS
The partial derivatives 𝑓𝑥 (𝑥, 𝑦) and 𝑓𝑦 (𝑥, 𝑦) represent the rates of change of 𝑓(𝑥, 𝑦)
in directions parallel to the x- and y-axes. In this section we will investigate rates
of change of 𝑓(𝑥, 𝑦) in other directions.

Figure 13.6.1 Figure 13.6.2

Geometrically, 𝑫𝒖 𝑓(𝑥0 , 𝑦0 ) can be interpreted as the slope of the surface 𝒛 =

𝒇 (𝒙, 𝒚) in the direction of u at the point (𝑥0 , 𝑦0 , 𝑓(𝑥0 , 𝑦0 )) (Figure 13.6.2).
Usually the value of 𝑫𝒖 𝑓(𝑥0 , 𝑦0 ) will depend on both the point (𝑥0 , 𝑦0 ) and the
direction u. Thus, at a fixed point the slope of the surface may vary with the
direction (Figure 13.6.3). Analytically, the directional derivative represents the
instantaneous rate of change of f (x, y) with respect to distance in the direction
of u at the point (𝑥0 , 𝑦0 ).
Example 1

Solution Using (2),

moved.
Example 2
Find the directional derivative of 𝑓(𝑥, 𝑦) = 𝑒 𝑥𝑦 at (−2, 0) in the direction
of the unit vector that makes an angle of π/3 with the positive x-axis.

Solution. The partial derivatives of f are

Example 3
Find the directional derivative of 𝑓 (𝑥, 𝑦, 𝑧) = 𝑥 2 𝑦 − 𝑦𝑧 3 + 𝑧 at the point
(1, −2, 0) in the direction of the vector a = 2i + j − 2k.
Solution.
Example 4
Let (𝑥, 𝑦) = 𝑥 2 𝑒 𝑦 . Find the maximum value of a directional derivative
at (−2, 0), and find the unit vector in the direction in which the maximum value
occurs.

Solution.
13.7 TANGENT PLANES AND NORMAL VECTORS

Example 1 Consider the ellipsoid 𝑥 2 + 4𝑦 2 + 𝑧 2 = 18.

(a) Find an equation of the tangent plane to the ellipsoid at the point (1, 2, 1).
(b) Find parametric equations of the line that is normal to the ellipsoid at the point
(1, 2, 1).
(c) Find the acute angle that the tangent plane at the point (1, 2, 1) makes with the
xy-plane.

Solution (a). We apply Definition 13.7.1 with 𝐹(𝑥, 𝑦, 𝑧) = 𝑥 2 + 4𝑦 2 + 𝑧 2

and (𝑥0 , 𝑦0 , 𝑧0 ) = (1, 2, 1). Since
Solution (b). Since n = < 2, 16, 2 > at the point (1, 2, 1), it follows from (4)
that parametric equations for the normal line to the ellipsoid at the point (1, 2, 1)
are x = 1 + 2t, y = 2 + 16t, z = 1 + 2t

Solution (c). To find the acute angle θ between the tangent plane and the xy-
plane, we

TANGENT PLANES TO SURFACES OF THE FORM 𝒛 = 𝒇 (𝒙, 𝒚)

To find a tangent plane to a surface of the form 𝑧 = 𝑓(𝑥, 𝑦), we can use
Equation (3) with the function 𝐹(𝑥, 𝑦, 𝑧) = 𝑧 − 𝑓(𝑥, 𝑦).

Example 2 Find an equation for the tangent plane and parametric equations for
the normal line to the surface 𝑧 = 𝑥 2 𝑦 at the point (2, 1, 4).

Solution. Let 𝐹(𝑥, 𝑦, 𝑧) = 𝑧 − 𝑥 2 𝑦. Then 𝐹(𝑥, 𝑦, 𝑧) = 0 on the surface, so we

can find the find the gradient of F at the point (2, 1, 4):
EXERCISE SET 13.7
3–12 Find an equation for the tangent plane and parametric equations for
the normal line to the surface at the point P.
13.8 MAXIMA AND MINIMA OF FUNCTIONS OF TWO VARIABLES
EXTREMA
If we imagine the graph of a function f of two variables to be a mountain range
(Figure 13.8.1), then the mountaintops, which are the high points in their
immediate vicinity,

Figure 13.8.1
are called relative maxima of f , and the valley bottoms, which are the low points
in their immediate vicinity, are called relative minima of f.

Just as a geologist might be interested in finding the highest mountain and deepest
valley in an entire mountain range, so a mathematician might be interested in
finding the largest and smallest values of f(x, y) over the entire domain of f. These
are called the absolute maximum and absolute minimum values of f. The following
definitions make these informal ideas precise.
FINDING RELATIVE EXTREMA
Recall that if a function g of one variable has a relative extremum at a point 𝑥0
where g is differentiable, then 𝑔(𝑥0 ) = 0. To obtain the analog of this result for
functions of two variables, suppose that 𝑓(𝑥, 𝑦) has a relative maximum at a point
(𝑥0 , 𝑦0 ) and that the partial derivatives of f exist at (𝑥0 , 𝑦0 ). It seems plausible
geometrically that the traces of the surface z = f(x, y) on the planes 𝑥 = 𝑥0 and
𝑦 = 𝑦0 have horizontal tangent lines at (𝑥0 , 𝑦0 ) (Figure 13.8.4), so 𝑓𝑥 (𝑥0 , 𝑦0 ) =
0 and 𝑓𝑦 (𝑥0 , 𝑦0 ) = 0.
The same conclusion holds if f has a relative minimum at (𝑥0 , 𝑦0 ), all of which
suggests the following result, which we state without formal proof.
 Figure 13.8.4

13.8.4 Theorem If f has a relative extremum at a point (𝑥0 , 𝑦0 ), and if the first
order partial derivatives of f exist at this point, then
𝑓𝑥 (𝑥0 , 𝑦0 ) = 0 and 𝑓𝑦 (𝑥0 , 𝑦0 ) = 0.

13.8.5 Definition A point (𝑥0 , 𝑦0 ) in the domain of a function 𝑓(𝑥, 𝑦) is called a

critical point of the function if 𝑓𝑥 (𝑥0 , 𝑦0 ) = 0 and 𝑓𝑦 (𝑥0 , 𝑦0 ) = 0 or if one or
both partial derivatives do not exist at (𝑥0 , 𝑦0 ).
Example 3

Example 4 Locate all relative extrema and saddle points of

𝑓(𝑥, 𝑦) = 4𝑥𝑦 − 𝑥 4 − 𝑦 4
Solution. Since
At the points (1, 1) and (−1, −1), we have 𝐷 > 0 and 𝑓𝑥𝑥 < 0, so relative
maxima occur at these critical points. At (0, 0) there is a saddle point since
𝐷 < 0.
9–20 Locate all relative maxima, relative minima, and saddle
points, if any.