The Tangent Plane and Normal Vectors to Surfaces
What you'll learn: How to compute tangent vectors and normal vectors to parametric surfaces using partial derivatives.
Core Idea
When you have a parametric surface defined by r(u,v) = ⟨x(u,v), y(u,v), z(u,v)⟩, you can find vectors that lie tangent to the surface at any point by taking partial derivatives with respect to the parameters.
Think of a parametric surface like a flexible sheet stretched in 3D space. At any point on this sheet, you can move in two independent directions—one by varying u while holding v constant, and another by varying v while holding u constant. These two directions give you tangent vectors.
Computing Tangent Vectors
The two fundamental tangent vectors at a point are:
- r_u = ∂r/∂u = ⟨∂x/∂u, ∂y/∂u, ∂z/∂u⟩
- r_v = ∂r/∂v = ⟨∂x/∂v, ∂y/∂v, ∂z/∂v⟩
These vectors span the tangent plane at that point.
Finding the Normal Vector
Since r_u and r_v lie flat against the surface, their cross product points perpendicular to both—giving you the normal vector:
n = r_u × r_v
This normal vector is perpendicular to the tangent plane and "sticks out" from the surface.
Example Analogy
Imagine you're standing on a hillside. Your two feet can slide in two independent directions along the slope—those are like your tangent vectors. A flagpole planted perpendicular to the ground at your feet represents the normal vector.
Key Takeaway: The partial derivatives r_u and r_v give tangent vectors to a parametric surface, and their cross product r_u × r_v produces the normal vector perpendicular to the surface.