Sign in to save

Bookmark this page so you can find it later.

Sign in to save

Bookmark this page so you can find it later.

Tangent Planes and Linearization of Surfaces cheat sheet - grade college

Click image to open full size

Calculus Grade college

Tangent Planes and Linearization of Surfaces Cheat Sheet

A printable reference covering tangent plane equations, linearization, differentials, gradients, and error approximation for college calculus.

TopicsDerivativesFunctionsVectorsGraphsCalculusStudy Guide
Download PNG

Related Tools

Related Labs

Related Worksheets

Related Infographics

Study as Flashcards

Tangent planes and linearization help you approximate a surface near a known point using a plane. This cheat sheet covers how to build tangent planes for surfaces written as z=f(x,y)z=f(x,y) and as level surfaces F(x,y,z)=cF(x,y,z)=c. These tools are essential in multivariable calculus because they connect partial derivatives, gradients, and local approximation.

They also support applications in error estimation, optimization, and numerical modeling.

The key idea is that a differentiable surface looks almost flat when you zoom in near a point. For z=f(x,y)z=f(x,y), the tangent plane uses the partial derivatives fx(a,b)f_x(a,b) and fy(a,b)f_y(a,b) as slopes in the xx and yy directions. The linearization L(x,y)L(x,y) is the tangent plane written as an approximation formula for f(x,y)f(x,y).

For an implicit surface F(x,y,z)=cF(x,y,z)=c, the gradient ∇F(a,b,c)\nabla F(a,b,c) is normal to the tangent plane.

Key Facts

  • For z=f(x,y)z=f(x,y), the tangent plane at (a,b,f(a,b))(a,b,f(a,b)) is z=f(a,b)+fx(a,b)(x−a)+fy(a,b)(y−b)z=f(a,b)+f_x(a,b)(x-a)+f_y(a,b)(y-b).
  • The linearization of f(x,y)f(x,y) at (a,b)(a,b) is L(x,y)=f(a,b)+fx(a,b)(x−a)+fy(a,b)(y−b)L(x,y)=f(a,b)+f_x(a,b)(x-a)+f_y(a,b)(y-b).
  • Near (a,b)(a,b), the approximation is f(x,y)≈L(x,y)f(x,y)\approx L(x,y) when ff is differentiable and (x,y)(x,y) is close to (a,b)(a,b).
  • The total differential is df=fx(a,b) dx+fy(a,b) dydf=f_x(a,b)\,dx+f_y(a,b)\,dy, where dx=x−adx=x-a and dy=y−bdy=y-b.
  • For an implicit surface F(x,y,z)=cF(x,y,z)=c, the tangent plane at (a,b,c)(a,b,c) is Fx(a,b,c)(x−a)+Fy(a,b,c)(y−b)+Fz(a,b,c)(z−c)=0F_x(a,b,c)(x-a)+F_y(a,b,c)(y-b)+F_z(a,b,c)(z-c)=0.
  • The gradient vector ∇F(a,b,c)=⟨Fx(a,b,c),Fy(a,b,c),Fz(a,b,c)⟩\nabla F(a,b,c)=\langle F_x(a,b,c),F_y(a,b,c),F_z(a,b,c)\rangle is perpendicular to the level surface F(x,y,z)=cF(x,y,z)=c.
  • A tangent plane to z=f(x,y)z=f(x,y) exists when fxf_x and fyf_y exist and are continuous near (a,b)(a,b).
  • The approximation error usually becomes smaller as (x,y)(x,y) gets closer to (a,b)(a,b), but linearization is only a local approximation.

Vocabulary

Tangent plane
A plane that best approximates a smooth surface near a point using the surface value and local slopes.
Linearization
The linear function L(x,y)L(x,y) that uses partial derivatives at a point to approximate f(x,y)f(x,y) nearby.
Partial derivative
A derivative such as fxf_x or fyf_y that measures how a function changes with respect to one variable while holding the other variables constant.
Total differential
The expression df=fx dx+fy dydf=f_x\,dx+f_y\,dy that estimates the small change in ff from small changes in xx and yy.
Gradient
The vector ∇F=⟨Fx,Fy,Fz⟩\nabla F=\langle F_x,F_y,F_z\rangle that points in the direction of greatest increase and is normal to a level surface.
Level surface
A surface defined by an equation F(x,y,z)=cF(x,y,z)=c, where the function FF has a constant value.

Common Mistakes to Avoid

  • Using fx(x,y)f_x(x,y) and fy(x,y)f_y(x,y) instead of fx(a,b)f_x(a,b) and fy(a,b)f_y(a,b) in the tangent plane. The slopes must be evaluated at the base point before writing the plane.
  • Forgetting the point shift terms (x−a)(x-a) and (y−b)(y-b). The tangent plane must be anchored at the point of tangency, not just written with xx and yy alone.
  • Mixing explicit and implicit formulas. Use z=f(a,b)+fx(a,b)(x−a)+fy(a,b)(y−b)z=f(a,b)+f_x(a,b)(x-a)+f_y(a,b)(y-b) for z=f(x,y)z=f(x,y), but use Fx(x−a)+Fy(y−b)+Fz(z−c)=0F_x(x-a)+F_y(y-b)+F_z(z-c)=0 for F(x,y,z)=cF(x,y,z)=c.
  • Assuming a tangent plane exists just because partial derivatives exist at one point. Differentiability is the condition that guarantees the plane gives a valid local linear approximation.
  • Using linearization far from the base point. The formula f(x,y)≈L(x,y)f(x,y)\approx L(x,y) is designed for nearby inputs, so accuracy can decrease quickly as the input moves away.

Practice Questions

  1. 1 Find the tangent plane to z=x2+3xy−y2z=x^2+3xy-y^2 at (1,2)(1,2).
  2. 2 Use linearization to approximate f(1.02,1.98)f(1.02,1.98) for f(x,y)=x2+3xy−y2f(x,y)=x^2+3xy-y^2 using the base point (1,2)(1,2).
  3. 3 Find the tangent plane to the level surface x2+y2+z2=14x^2+y^2+z^2=14 at (1,2,3)(1,2,3).
  4. 4 Explain why the gradient ∇F(a,b,c)\nabla F(a,b,c) is used as a normal vector when finding the tangent plane to F(x,y,z)=cF(x,y,z)=c.

Understanding Tangent Planes and Linearization of Surfaces

A tangent plane records first order change. This means it keeps the part of a surface's behavior that changes at a steady rate near the chosen point. The partial derivative in one input direction tells how the output changes when only that input moves.

The other partial derivative does the same job in the second direction. Together, these rates describe every small horizontal move, not just moves parallel to the coordinate axes. A move that goes diagonally can be separated into two component moves.

The predicted height change is the sum of the changes from those components. This is why partial derivatives are more useful together than separately.

For an implicitly defined surface, the gradient gives a different viewpoint. It points in the direction where the defining quantity increases most quickly. Moving along the surface does not change that quantity, so every direction contained in the tangent plane must be perpendicular to the gradient.

This fact is especially helpful for spheres, ellipsoids, and surfaces that are difficult to solve for height. At a point on a sphere, the gradient points directly outward from its center. The tangent plane is therefore perpendicular to the radius through that point.

A usable tangent plane requires a nonzero gradient. If the gradient is zero, the surface may have a singular point, a crossing, or no single well-defined tangent plane.

The differential measures a predicted small output change rather than the output itself. Its units matter. If the inputs are measured in meters and the output is measured in liters, each partial derivative has units of liters per meter.

Multiplying each rate by a small input change produces liters, so the combined differential has the correct output unit. This makes differentials useful in measurement work. Small uncertainty in a radius, length, or temperature can be converted into an estimated uncertainty in volume, area, or energy.

The estimate can be positive or negative depending on the direction of change. For maximum possible error, students often use absolute values so that separate contributions cannot cancel by accident.

Linearization becomes less reliable when the surface bends strongly or when the input move is not small. Second partial derivatives describe this bending, often called curvature in this setting. If the slopes themselves change rapidly, a plane leaves the surface quickly.

A practical habit is to choose a nearby base point where the function value and partial derivatives are easy to calculate. Then compare the input changes with the scale of the problem. Keep track of the base point, since derivatives evaluated at a different point give a different plane.

For calculator or computer results, test the approximation at a point very close to the base point first. A small error there is reassuring, while a large error usually signals an arithmetic mistake, a wrong derivative, or a point that is too far away.