Examples on Tangent Planes and Normal Lines
For the interest of lecture time, I would like to give you a few examples of computing tangent
planes and normal lines here. Please read and study these examples!
Example 1: Find the equations of the tangent plane and the normal line at P0 (−2, 1, −3)
to the ellipsoid:
x2 z2
+ y2 + = 3. (1)
4 9
Solution: First of all, the equation of an ellipsoid (close to a rugby ball shape) with radii
α, β, γ along the x, y, z axes with the origin as a center is:
x2 y2 z2
+ + = 1.
α2 β 2 γ 2
You can think this as a generalization (in fact, a stretched and squeezed version) of the unit
sphere x2 + y 2 + z 2 = 1. See Figure 1.
Hence, Eq. (1) is also an ellipsoid because it
is the same as:
x2 y 2 z 2
+ + = 1. 5
12 3 27
Let’s go back to Equation (1) now. This is a
level surface f (x, y, z) = c where f (x, y, z) is 0
z
the lefthand side of Eq.(1) and c = 3. In gen-
eral, let P (x, y, z) be any point on the plane
tangent to the level surface f (x, y, z) = c at −5
5
P0 (x0 , y0 , z0 ). As we discussed in my lecture, 5
0
the gradient vector ∇f at P0 is perpendicular 0
to this tangent plane, i.e., y −5 −5
x
−−→ x2 y2 z2
∇f (P0 ) P0 P = 0, Figure 1: The ellipsoid 12
+ 3
+ 27
= 1.
which can be written as
fx (x0 , y0 , z0 ) · (x − x0 ) + fy (x0 , y0 , z0 ) · (y − y0 ) + fz (x0 , y0 , z0 ) · (z − z0 ) = 0. (2)
Hence, in this particular case,
x 2
∇f (x, y, z) = , 2y, z .
2 9
So,
2
∇f (P0 ) = ∇f (−2, 1, −3) = −1, 2, − .
3
1
Finally, the equation of the tangent plane at P0 can be written as
2
−1 · (x + 2) + 2 · (y − 1) − (z + 3) = 0,
3
which can be written in a nice form as:
3x − 6y + 2z + 18 = 0 .
As for the normal line at P0 , we use the following formula:
x = x0 + fx (P0 ) · t, y = y0 + fy (P0 ) · t, z = z0 + fz (P0 ) · t, (3)
where t is an arbitrary real-valued parameter. So, in this particular case, we have.
2
x = −2 − t, y = 1 + 2t, z = −3 − t ,
3
or without using t (review Section 12.5!),
x+2 y−1 z+3
= = .
−1 2 − 23
Example 2: Find the tangent plane and the normal line at P0 (1, 1, −1) to the following
surface:
z = x2 − xy − y 2 (4)
Solution: Eq. (4) is the same as:
x2 − xy − y 2 − z = 0.
So, f (x, y, z) = x2 − xy − y 2 − z and c = 0. Now, simply follow the formulas as before. Please
do it yourself to confirm: For the tangent plane:
x − 3y − z + 1 = 0 .
For the normal line:
x = 1 + t, y = 1 − 3t, z = −1 − t ,
or if you do not want to use t,
x−1 y−1 z+1
= = .
1 −3 −1
In general, the tangent plane and normal line at P0 (x0 , y0 , z0 ) to a surface z = f (x, y)
can be computed as follows. Let F (x, y, z) = f (x, y) − z = 0. Now, we can use the previous
formulas Equations (2), (3) by replacing f by F .
∇F = hFx , Fy , Fz i
= hfx , fy , −1i .
2
Note Fz = −1 here, which is obvious from the definition of F (x, y, z). Hence, the gradient
vector of F at P0 is:
∇F (P0 ) = hfx (P0 ), fy (P0 ), −1i .
Now, using Eq. (2), the tangent plane becomes:
fx (P0 ) · (x − x0 ) + fy (P0 ) · (y − y0 ) − 1 · (z − z0 ) = 0.
Furthermore, since P0 is on the surface, so z0 = f (x0 , y0 ) = f (P0 ). Finally, we have
fx (P0 ) · (x − x0 ) + fy (P0 ) · (y − y0 ) − z + f (P0 ) = 0,
or
z = f (P0 ) + fx (P0 ) · (x − x0 ) + fy (P0 ) · (y − y0 ).
As for the normal line, from Eq. (3), we have:
x = x0 + fx (P0 ) · t, y = y0 + fy (P0 ) · t, z = f (P0 ) − t,
or
x − x0 y − y0 z − f (P0 )
= = .
fx (P0 ) fy (P0 ) −1
Of course, in the above formulas, you can use z0 instead of f (P0 ).
