Partial Derivatives

Tangent Planes, Linear Approximations

Partial Derivatives

Math 301
Lecture 3

Rami Younes Partial Derivatives

 Partial Derivatives
Tangent Planes, Linear Approximations

Lecture Outline

1 Partial Derivatives
Partial Derivatives of Functions of Two Variables
Interpretation of Partial Derivatives
Partial Derivatives of Functions in any number of variables

2 Tangent Planes, Linear Approximations

Tangent Planes
Linear Approximations

Rami Younes Partial Derivatives

 Partial Derivatives
Tangent Planes, Linear Approximations

Recommended Reading

Stewart’s Caluculus (6th edition): 2.7 (optional), the titles

covered in sections 14.3, 14.4.
Thomas’ Caluculus (13th edition): section 14.3 and 14.6 (only
the titles in this lecture)

Rami Younes Partial Derivatives

 Partial Derivatives of Functions of Two Variables
Partial Derivatives
Interpretation of Partial Derivatives
Tangent Planes, Linear Approximations
Partial Derivatives of Functions in any number of variables

First Order Partial Derivatives

Recall Example
If g is a function of a real variable then the derivative of g at Let f (x, y ) = x 3 + x 2 y 3 2y 2 .
a is
d d 3 2
0 g (a + h) g (a) fx (2, 1) = f (x, 1) = (x + x 2) = 16.
g (a) = lim . dx x=2 dx x=2
h!0 h d d 3 2
fy (2, 1) = f (2, y ) = (8 + 4y 2y ) = 8.
Partial derivatives of two variable functions dy y =1 dy y =1
Let f (x, y ) be a function of two variables defined in an open
ball that contains the point (a, b). Remark
More generally, if (x, y ) is an arbitrary point
Fix the value of y = b to obtain a function of a real
variable x, namely, g (x) = f (x, b). fixed
@f d z}|{ f (x + h, y ) f (x, y )
(x, y ) = f (x, y ) = lim
The partial derivative of f with respect to x at @x dx h!0 h
@f (a, b) or f (a, b), is the
(a, b), denoted @x fixed
x @f d z}|{ f (x, y + h) f (x, y )
derivative of g at x = a: (x, y ) = f ( x , y ) = lim
@y dy h!0 h
@f 0 f (a + h, b) f (a, b)
(a, b) = g (a) = lim
@x h!0 h Example
Let f (x, y ) = x 3 + x 2 y 3 2y 2 .
Fix the value of x = a to obtain a function of a real
variable y , namely, G (y ) = f (a, y ). @f
hold y fixed and di↵
z}|{ 2 3
The partial derivative of f with respect to y at (x, y ) = 3x + 2xy
@f (a, b) or f (a, b), is the @x
(a, b), denoted @y y hold x fixed and di↵
@f z}|{ 2 2
derivative of G at y = b: (x, y ) = 3x y 4y
@y
@f 0 f (a, b + h) f (a, b)
(a, b) = G (b) = lim
@y h!0 h
Rami Younes Partial Derivatives
 Partial Derivatives of Functions of Two Variables
Partial Derivatives
Interpretation of Partial Derivatives
Tangent Planes, Linear Approximations
Partial Derivatives of Functions in any number of variables

First Order Partial Derivatives

Remarks 2y
2 Find fx as a function if f (x, y ) = y +cos x
.
Partial derivatives may or may not exist (being limits).
If we let (x, y ) vary, then fx (x, y ) and fy (x, y ) are ✓ ◆
@ 2y
functions of two variables. fx =
@x y + cos x
Notation for Partial Derivatives
If z = f (x, y ), we write quotient rule @ (2y ) @ (y + cos x)
z}|{ (y + cos x) @x 2y @x
=
(y + cos x)2
@f @ @z
fx (x, y ) = fx = = f (x, y ) = = f 1 = D1 f = Dx f 2y sin x
@x @x @x =
(y + cos x)2
@f @ @z
fy (x, y ) = fy = = f (x, y ) = = f 2 = D2 f = Dy f
@y @y @y ⇣ ⌘
3 If f (x, y ) = sin x , calculate the partial
1+y
derivatives of f .
Examples
1 Find fy as a function if f (x, y ) = y sin xy . ✓ ◆ ✓ ◆
@f chain
z}|{
rule
x @ x
= cos
@x 1+y @x 1+y
@
product rule
@ @ ✓ ◆
z}|{ x 1
fy = (y sin xy ) = (y ) sin xy + y (sin xy ) = cos
@y @y @y 1+y 1+y
= sin xy + xy cos xy ✓ ◆ ✓ ◆
|{z} @f chain
z}|{
rule
x @ x
chain rule = cos
@y 1+y @y 1+y
✓ ◆
x x
= cos
1+y (1 + y )2

Rami Younes Partial Derivatives

 Partial Derivatives of Functions of Two Variables
Partial Derivatives
Interpretation of Partial Derivatives
Tangent Planes, Linear Approximations
Partial Derivatives of Functions in any number of variables

Higher Order Partial Derivatives

Second Partial Derivatives
If f is a function of two variables, the partial derivatives of f , fx @f @f
2 3 2 2
and fy , are also functions of two variables, so their partial (x, y ) = 3x + 2xy and (x, y ) = 3x y 4y
derivatives may be considered and are called the second partial @x @y
derivatives of f :
Therefore,
(fx )x , (fx )y and (fy )x , (fy )y .
@ 2 3 3
fxx = (3x + 2xy ) = 6x + 2y
The following notation is used @x
@ 2 2 2
✓ ◆ 2
fyy = (3x y 4y ) = 6x y 4
@ @f @ f @y
(fx )x = fxx = f11 = =
@x @x @x 2 @ 2 3 2
✓ ◆ fxy = (3x + 2xy ) = 6xy
2 @y
@ @f @ f
(fx )y = fxy = f12 = =
@y @x @y @x @ 2 2 2
fyx = (3x y 4y ) = 6xy
✓ ◆ @x
@ @f @2f
(fy )x = fyx = f21 = =
@x @y @x@y
Remark
✓ ◆ 2
@ @f @ f In the preceding example, we obtained fxy = fyx . It is not a
(fy )y = fyy = f22 = = coincidence! The following theorem holds:
@y @y @y 2
Clairaut’s Theorem
Suppose that f is a function of two variables that is defined
Example in an open disk D. If the functions fxy and fyx are both
Find the second partial derivatives of f (x, y ) = x 3 + x 2 y 3 2y 2 . continuous on D, then
Solution
It has been found earlier that: fxy (x, y ) = fyx (x, y )

Rami Younes Partial Derivatives

 Partial Derivatives of Functions of Two Variables
Partial Derivatives
Interpretation of Partial Derivatives
Tangent Planes, Linear Approximations
Partial Derivatives of Functions in any number of variables

Algebraic Interpretation of Partial Derivatives

Let z = f (x, y ).
y fixed
Solution
@f z}|{ d (a)
(x, y ) = f (x, y ), thus:
@x dx
@f @T
is the instantaneous rate of change of f w.r.t. x = T1 e
x
sin(!t x) T1 e
x
cos(!t x)
@x @x
while y is held fixed.
x
x fixed = T1 e (sin(!t x) + cos(!t x))
@f z}|{ d
(x, y ) = f (x, y ), thus:
@y dy
Interpretation: on a given day (as indicated by the
@f fixed t), the instantaneous rate of change of the
is the instantaneous rate of change of f w.r.t. y
@y @T
temperature at a given depth x is (x, t).
while x is held fixed. @x
(b)
Example
In a study of frost penetration it was found that the
temperature T at time t (measured in days) at a depth x @T x
(measured in feet) can be modeled by the function = !T1 e cos(!t x)
@t

x
T (x, t) = T0 + T1 e sin(!t x) Interpretation: at a given depth (as indicated by the
fixed x), the rate of change of the temperature on a
where ! = 2⇡/365 and is a positive constant. @T
given day t (as time flows) is (x, t).
@T @t
(a) Find . What is its physical interpretation?
@x
@T
(b) Find . What is its physical interpretation?
@t

Rami Younes Partial Derivatives

 Partial Derivatives of Functions of Two Variables
Partial Derivatives
Interpretation of Partial Derivatives
Tangent Planes, Linear Approximations
Partial Derivatives of Functions in any number of variables

Geometric Interpretation of Partial Derivatives

The definition of partial derivatives
Let f (x, y ) be a function of two variables defined in an open
ball that contains the point (a, b).
@f 0 @f 0
Geometric Interpretation (a, b) = g (a) and (a, b) = G (b)
@x @y
@f d
(a, b) = f (x, b) implies that
@x dx x=a
@f d @f
(a, b) = f (a, y ) (a, b) is the slope of the tangent line T1 to C1
@y dy y =b @x
at P(a, b, c), (where c = f (a, b)).
Note that: @f
(a, b) is the slope of the tangent line T2 to C2
the graph C1 of the function g (x) = f (x, b) is @y
obtained by slicing the graph z = f (x, y ) by the at P(a, b, c).
vertical plane y = b
the graph C2 of the function G (y ) = f (a, y ) is Remarks
obtained by slicing the graph z = f (x, y ) by the (i) Recall that in the plane, if the slope of a straight
vertical plane x = a line is m, then a direction vector of the line is (1, m)
(ii) It follows from remark (i) that a vector in the
direction of the tangent line T1 is
@f )
v1 = (1, 0, @x
(iii) It follows from remark (i) that a vector in the
direction of the tangent line T2 is
@f )
v2 = (0, 1, @y

Rami Younes Partial Derivatives

 Partial Derivatives of Functions of Two Variables
Partial Derivatives
Interpretation of Partial Derivatives
Tangent Planes, Linear Approximations
Partial Derivatives of Functions in any number of variables

Partial Derivatives of Functions in any number of variables

Functions of more than two variables Holding x and z fixed, then di↵erentiating w.r.t y gives:
Partial derivatives can also be defined for functions of three or
more variables. For example, if f is a function of three variables x,
y , and z, then its partial derivative with respect to x is defined as fy = xe xy ln z.

Holding x and y fixed, then di↵erentiating w.r.t z gives:

fixed fixed
@f d z}|{ z}|{
(x, y , z) = f (x, y , z ) xy
@x dx fz = e z .
f (x + h, y , z) f (x, y , z)
= lim
h!0 h @u
Interpretation of
@xi
@u
More generally, if u = f (x1 , x2 , . . . , xn ) is a function of n is the instantaneous rate of change of u w.r.t. xi when
variables, its partial derivative with respect to the i th variable xi is @xi
found by holding all other variables fixed and by di↵erentiating all other variables are held fixed.
with respect to xi . We also write Remark
Second partial derivatives can be found for functions of
@u @f several variables regardless of the number of variables.
= = f x i = Di f = Dx i f
@xi @xi Example
Let f (x, y , z) = sin(3x + yz). Find fyz .
Example Solution
Find the partial derivatives of f (x, y , z) = e xy ln z.
Solution @
Holding y and z fixed, then di↵erentiating w.r.t x gives: fyz = (fy )z = (z cos(3x + yz))
@z
= cos(3x + yz) yz sin(3x + yz)
fx = ye xy ln z.

Rami Younes Partial Derivatives

 Partial Derivatives of Functions of Two Variables
Partial Derivatives
Interpretation of Partial Derivatives
Tangent Planes, Linear Approximations
Partial Derivatives of Functions in any number of variables

Partial Derivatives of Functions in any number of variables

Example
Let f (x, y , z) = sin(3x + yz). Find fzy . Remark
Partial derivatives of order 3 or higher can also be defined.
Solution For instance,
!
@ @ @ @2f @3f
fzy = (fz ) = (y cos(3x + yz)) fxyy = (fxy )y = = .
@y @y @y @y @x @y 2 @x
= cos(3x + yz) yz sin(3x + yz) = fyz
Example
Remark Let f (x, y , z) = sin(3x + yz). Calculate fxxyz .
Notice that for the function in the preceding example,
fyz = fzy . In fact, it is straight forward to check that also fx = 3 cos(3x + yz)
fxx = 9 sin(3x + yz)
fxz = fzx and fxy = fyx .
fxxy = 9z cos(3x + yz)
Again, this is not a coincedence. fxxyz = 9 cos(3x + yz) + 9yz sin(3x + yz)
Remark
If the second partial derivatives of a function of several
variables u = f (x1 , x2 , . . . , xn ) are continuous then
Clairaut’s Theorem will still hold, namely,

@2u @2u
=
@xi @xj @xj @xi

for any i 6= j with i and j in {1, 2, . . . , n}.

Rami Younes Partial Derivatives

 Partial Derivatives Tangent Planes
Tangent Planes, Linear Approximations Linear Approximations

Tangent Plane to a Surface

Suppose that a surface has an equation z = f (x, y )
and let P(a, b, f (a, b)) be a point on the surface. Remark
Assume that the surface is smooth, that is, it has no (i) A vector in the direction of the tangent line T1 is
sharp peaks (which is guaranteed if fx and fy are
v1 = (1, 0, fx (a, b))
continuous).
(ii) A vector in the direction of the tangent line T2 is
Tangent Plane
v2 = 0, 1, fy (a, b)
Let C1 and C2 be the curves obtained by
slicing the surface by the vertical planes Equation of the Tangent Plane
y = b and x = a respectively. A normal vector to the tangent plane at P is
Let T1 and T2 be the tangent lines to C1 and
C2 at the point P. i j k
n = v1 ⇥ v 2 = 1 0 fx = fx , fy , 1
0 1 fy

Therefore, if M(x, y , z) is an arbitrary point, the equation of the plane is

given by

!
PM.n = 0 , fx (a, b)(x a) fy (a, b)(y b) + z f (a, b) = 0

Therefore,

z = f (a, b) + fx (a, b)(x a) + fy (a, b)(y b)

The tangent plane to the surface at P is the plane
defined by the lines T1 and T2 .

Rami Younes Partial Derivatives

 Partial Derivatives Tangent Planes
Tangent Planes, Linear Approximations Linear Approximations

Linear Approximations
Linear Approximations
Example The linear function
Find the tangent plane to the paraboloid z = 2x 2 + y 2 at
the point (1, 1, 3).
L(x, y ) = f (a, b) + fx (a, b)(x a) + fy (a, b)(y b)
Solution
f (x, y ) = 2x 2 + y 2 . Then whose graph is the tangent plane at P(a, b, f (a, b)) is called the
linearization of f at (a, b).
fx (x, y ) = 4x fy (x, y ) = 2y For (x, y ) near (a, b), f (x, y ) ⇡ L(x, y ).
fx (1, 1) = 4 fy (1, 1) = 2 Example
Find the linearization of f (x, y ) = xe xy at (1, 0) then use it to
The equation of the tangent plane at (1, 1, 3) is estimate f (1.1, 0.1).
z = 3 + 4(x 1) + 2(y 1) , z = 4x + 2y 3.
Solution
Remark
The tangent plane provides a good approximation of the fx (x, y ) = e
xy
+ xye
xy
fy (x, y ) = x e
2 xy
surface near the point of tangency.
fx (1, 0) = 1 fy (1, 0) = 1
In the above example, gradually zooming-in towards the
point (1, 1, 3) the surface looks increasingly flatter,
resembling its tangent plane. The linearization is

L(x, y ) = f (1, 0) + fx (1, 0)(x 1) + fy (1, 0)(y 0) = x + y .

Therefore, xe xy ⇡ x + y for (x, y ) near (1, 0).

f (1.1, 0.1) ⇡ L(1.1, 0.1) = 1.1 0.1 = 1.

The actual value of f (1.1, 0.1) = 1.1e 0.11 = 0.98542 . . .

Rami Younes Partial Derivatives